text
stringlengths 14
5.77M
| meta
dict | __index_level_0__
int64 0
9.97k
⌀ |
|---|---|---|
Multicraft was added by Mastergalen in Jun 2012 and the latest update was made in Apr 2019. The list of alternatives was updated Mar 2019 There is a history of all activites on Multicraft in our Activity Log. It's possible to update the information on Multicraft or report it as discontinued, duplicated or spam.
Discontinued No longer under active development, but XereoNet is working on a new version called SpaceCP that will work with both Bukkit and Vanilla servers.
Discontinued Bukkit Webby has not been updated since February 23, 2014.
Alternatives to Multicraft for Linux, Windows, Mac, Self-Hosted, GitHub and more. Filter by license to discover only free or Open Source alternatives. This list contains a total of 7 apps similar to Multicraft.
No reviews yet for Multicraft, want to be first?
Maybe you want to be the first to submit a comment about Multicraft? Just click the button up to your right!
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,567
|
\section{Introduction} \label{intro}
We recently analyzed Gamow-Teller (GT) strengths in $^{48}$Ca and $^{78}$Ni with the self--consistent charge--exchange (CE) SSRPA \cite{gamba2020}, based on Skyrme effective interactions \cite{sk1,sk2,vautherin}, in the framework of energy--density--functional (EDF) theories.
A subtraction procedure, employed to handle instabilities, to avoid overcounted correlations, and to regularize ultraviolet divergences \cite{tse2013,gamba2015}, was fully applied by inverting the matrix $A_{22}$, which represents the two particle - two hole (2p2h) sector of the SSRPA matrix. Reference \cite{gamba2020} illustrates the first application of this procedure within a CE version of the SSRPA model.
The agreement of the predicted GT strength with the corresponding experimental distribution in $^{48}$Ca \cite{yako} was shown to be significantly improved, compared to other available theoretical predictions (see for example Refs. \cite{cao,niu}). In these models the Ikeda sum rule \cite{ike}, integrated up to an excitation energy of 20-30 MeV, is systematically overestimated and is predicted much closer to the full value $3(N-Z)$ than the experimental measurement. When this overestimation occurs, it is customary to
resort to {\it{ad-hoc}} quenching factors in the GT operator, which are extensively employed (both in mean-field and in beyond-mean-field approaches within EDF models) to improve the agreement with the experimental measurements.
For the first time, in Ref. \cite{gamba2020}, the experimental quenching of the Ikeda sum rule could be well accounted for in $^{48}$Ca without adopting {\it{ad-hoc}} factors.
This achievement might have implications for future applications of the CE-SSRPA model to the evaluation of the nuclear matrix elements (NMEs) entering in neutrinoless-double-beta ($0\nu \beta \beta$) decay half-lives, for example in $^{48}$Ca, which is expected to be the lightest $\beta \beta$ emitter.
The mechanisms underlying GT resonances and $\beta$ decays have a deeply different nature. Nevertheless,
a GT-type term turns out to be the leading contribution in $0\nu \beta \beta$-decay NMEs. In addition,
it is known that most of the theoretical models which resort to {\it ad-hoc} quenched GT operators also tend to overestimate single $\beta$-decay rates (compared to available data) and need, in practice, to quench by hand the weak interaction axial-vector coupling constant $g_A$ by a factor of $\approx$ 0.75 (see for instance Refs. \cite{towner,wilkinson,brown,chou,martinez}).
The impact that a quenched $g_A$ value would have in particular on the sensitivity of $0\nu\beta\beta$ measurements was analyzed in Ref. \cite{suhonen} and was shown to be far from being negligible.
For all these reasons,
a prediction of $0\nu\beta\beta$ NMEs (based on the bare value of the coupling constant $g_A$) would be significantly more reliable if obtained with a model where,
coherently, {\it{ad-hoc}} quenching procedures are not required to describe satisfactorily experimental GT
strengths and single $\beta$-decay rates. We also mention that another direction explored in the literature consists in evaluating rigorously the quenching of the transition operator by a proper computation of the correlations missing in the model as done, for example, in the shell-model calculations of Ref. \cite{coraggio}, based on realistic potentials and the many-body-perturbation-theory approach.
The low-energy part of the GT strength of $^{78}$Ni was used in Ref. \cite{gamba2020} to evaluate the
$\beta$-decay half-life in this nucleus (using the bare value of $g_A$). The comparison with the experimental value \cite{hosmer,david} was found to be quite good.
The success of the CE-SSRPA model was ascribed in Ref. \cite{gamba2020} to the explicit inclusion of 2p2h configurations. As a matter of fact, these configurations
have a density which increases with the excitation energy, producing a high--energy tail in the spectrum as was anticipated by some authors in the literature (see for instance Refs. \cite{bertsch,brink,arima}) but never demonstrated in practice before the work published in Ref. \cite{gamba2020}. Such a mechanism effectively pushes part of the strength towards higher excitation energies and leads to a better agreement with the reduced percentage of the Ikeda sum rule which is measured experimentally. What was called for many years ``the puzzle of the missing strength" could be finally elucidated.
In the present work, we aim to extend our analysis of GT strengths based on the CE-SSRPA model to other nuclei, all closed-shell or closed-subshell, spanning different regions of the nuclear chart.
The calculations performed for $^{48}$Ca in Ref. \cite{gamba2020} were carried out by applying a full subtraction procedure, that is, without resorting to a diagonal approximation in the $A_{22}$ matrix to invert. We show in the present work that this is indeed a crucial point in GT calculations for $^{48}$Ca, where the coupling between 2p2h configurations has to be entirely taken into account not only in the matrix to diagonalize but also in the matrix to invert for the subtraction.
Such fully implemented calculations require a considerable computational effort.
For this reason, a special care must be devoted to the numerical implementation of the subtraction procedure and we employ here a method based on a 2$\times$2 block matrix inversion that can be found in standard textbooks \cite{inversion1,inversion2,inversion3}.This procedure allows us to strongly reduce the computational effort to achieve the inversion, making it feasible also for medium-mass nuclei.
After this preliminary analysis on $^{48}$Ca, we examine two lighter nuclei, $^{14}$C and $^{22}$O (where a full inversion of the matrix $A_{22}$ may be carried out much more easily), in order to compare our results with the {\it{ab-initio}} coupled-cluster predictions of Ref. \cite{ekstrom}. Such a comparison is particularly interesting because, in this specific {\it{ab-initio}} scheme, 2p2h configurations are also included.
The authors of Ref. \cite{ekstrom} show in particular the importance of using, together with a a three-nucleon force, coherent two-body currents in the transition operator.
In our case, we employ effective Skyrme functionals. As far as three-body forces are concerned, these functionals contain effectively a three-body contribution through the density-dependent two-body term appearing in the interaction. On the other side, we systematically use one-body GT transition operators.
Hence, the comparison with the quenching of GT strengths predicted with the coupled-cluster approach will tell us whether, within our model, the inclusion of a two-body part in the transition operator is (or not) an important missing ingredient. If we find strengths which are similarly or more strongly quenched compared to those of Ref. \cite{ekstrom} for the same nuclei, we may conclude that the inclusion of a two-body transition operator is not crucial in our case.
We then treat two nuclei heavier than $^{48}$Ca, namely $^{90}$Zr and $^{132}$Sn, and compare strengths and integrated strengths with the available experimental data \cite{zr1,zr2,sn}. We check in particular whether, also in these cases as for $^{48}$Ca, the experimental quenching of the integrated strengths is better reproduced compared to other theoretical models.
The article is organized as follows. Section \ref{form} briefly provides details on the performed calculations and the used transition operator.
In Sec. \ref{48ca}, we discuss the GT strength in
$^{48}$Ca and show that a full inversion of the $A_{22}$ matrix is necessary.
We present in Sec. \ref{light} results for the nuclei $^{14}$C and $^{22}$O as well as the comparisons with Ref. \cite{ekstrom}. Section \ref{heavy} illustrates results for the heavier nuclei $^{90}$Zr and $^{132}$Sn and the comparisons with the corresponding experimental measurements, published in Refs. \cite{zr1,zr2} and \cite{sn}, respectively. Conclusions are drawn in Sec. \ref{conclu}.
\section{Details on the calculations} \label{form}
The reader may find details on the CE second-random-phase-approximation (CE-SRPA) formalism (without subtraction) in Ref. \cite{brink} and details on the SSRPA formalism (for charge-conserving excitations) in Ref. \cite{gamba2015}. The first CE-SSRPA results (that is, CE results obtained by applying a subtraction procedure within the SRPA scheme) were presented in Ref. \cite{gamba2020}, as already mentioned in Sec. \ref{intro}.
In this work, CE-SSRPA calculations are carried out on top of Hartree-Fock ground states, using the Skyrme parametrization SGII \cite{sgii1,sgii2}.
The cutoff on one particle-one hole (1p1h) configurations is set at 60 MeV, whereas the one on 2p2h configurations is in most cases chosen equal to a value between 40 and 50 MeV. We have checked that all the used cutoff values ensure that the relevant physics is always included leading in fact to cutoff independent results.
We employ GT one-body transition operators
\begin{equation}
\hat{O}^{\pm}=\sum_{i=1}^{A} \sum_{\mu} \sigma_{\mu}(i) \tau^{\pm}(i),
\label{oper}
\end{equation}
for a nucleus with $A$ nucleons; $\tau^{\pm}$ are the
isospin raising ($+$) and lowering ($-$) operators, $\tau^{\pm}=t_x \pm it_y$, and $\sigma_{\mu}$ is the spin operator.
The $\hat{O}^+$ operator generates the GT$^+$ strength (a neutron is added and a proton is removed), whilst the $\hat{O}^-$ operator produces the GT$^-$ strength (a neutron is removed and a proton is added).
The Ikeda GT sum rule \cite{ike}, relating the integrated strengths $S$ of the GT$^-$ and the GT$^+$ spectra to the number of neutrons $N$ and protons $Z$ of the nucleus, has the following expression:
\begin{equation}
S_{GT^-}-S_{GT^+}=3(N-Z).
\label{ikeda}
\end{equation}
It is known that the strength $S_{GT^-}$ is dominant in nuclei having a neutron excess.
This sum rule is model independent and can be easily deduced by using the properties of the isospin operators if the condition of completeness of states is fulfilled.
It was underlined in particular in Ref. \cite{gamba2020} that, within the random-phase approcimation (RPA) and extended-RPA models, this sum rule holds if the quasiboson approximation is used.
Since the inclusion of 2p2h configurations pushes a non negligible part of the SSRPA strength to high energies, the full value $3(N-Z)$ is reached only by integrating up to a very high excitation energies, much higher than in RPA. As already stressed, this mechanism allows for a strongly improved agreement of the integrated strength with the experimental cumulative Ikeda sum in the nucleus $^{48}$Ca.
\section{Diagonal approximation in $^{48}$Ca} \label{48ca}
The full GT$^-$ spectrum for $^{48}$Ca was published in Ref. \cite{gamba2020}, where the SGII Skyrme interaction was used. The same interaction is employed here. This spectrum corresponds to the SSRPAFF curve shown in
panel (a) of Fig. \ref{ffd}, where different GT$^-$ strengths are plotted. All the theoretical distributions are obtained by folding the discrete spectra with a Lorentzian having a width of 1 MeV. The experimental data are extracted from Ref. \cite{yako}. As indicated above, the spectrum obtained with a full calculation is denoted by the acronym SSRPAFF. The acronym SSRPADD refers to a calculation where the diagonal approximation is used twice for the matrix $A_{22}$, both in the inversion and in the diagonalization procedures. Finally, SSRPDF indicates a calculation where the diagonal approximation for the matrix $A_{22}$ is adopted only in the subtraction procedure, where this matrix is inverted, whereas $A_{22}$ is fully treated in the diagonalization. Also RPA results are shown for comparison.
Figure \ref{ffd} shows that the SSRPAFF model leads to a visible improvement of the results compared to RPA, as already discussed in Ref. \cite{gamba2020}. The double diagonal approximation (SSRPADD) provides a worse reproduction of the full spectrum, yet showing a considerable improvement with respect to RPA. The hybrid SSRPADF computation leads to better results, but still different from the SSRPAFF ones.
Whereas Ref. \cite{gamba2015} showed for charge-conserving excitations that hybrid ``DF"-type results were not strongly different compared to the full spectrum, Fig. \ref{ffd} clearly indicates that this is no longer the case here.
Panel (b) of Fig. \ref{ffd} confirms that the SSRPADF scheme reproduces sensibly less well than SSRPAFF the experimental strength and this is visible in the displayed cumulative sums. Such a result indicates that, in the calculations done for $^{48}$Ca, the inversion of the matrix in the subtraction procedure cannot be carried out by adopting a simple diagonal approximation in the matrix and needs a more accurate treatment.
\begin{figure}
\includegraphics[scale=0.32]{Ca48-SGII-ALL.eps}
\caption{(a) GT$^-$ strengths obtained with the Skyrme interaction SGII for the nucleus $^{48}$Ca in MeV$^{-1}$. Experimental data are extracted from Ref. \cite{yako}. RPA and all SSRPA strengths are obtained by folding with a Lorentzian having a width of 1 MeV. (b) Cumulative sum of the strength up to the excitation energy of 20 MeV. See text for more details.}
\label{ffd}
\end{figure}
\section{Lighter nuclei: $^{14}$C and $^{22}$O.} \label{light}
Before treating nuclei heavier than $^{48}$Ca, we address and check in lighter systems another aspect of our calculations.
We have chosen the nuclei $^{14}$C and $^{22}$O as illustrations, because the GT Ikeda sum rule was computed for these nuclei in Ref. \cite{ekstrom} with coupled-cluster calculations.
In these calculations based on effective-field theories \cite{epelbaum}
the GT transition operator, Eq. (\ref{oper}), is corrected by the inclusion of two-body currents, in a coherent way with the inclusion of three-body forces.
The impact of the inclusion of two-body currents in the transition operator was analyzed in Ref. \cite{ekstrom} by studying three nuclei, $^{14}$C, $^{22}$O, and $^{24}$O. The authors found that such a modification of the operator induces a change in the total Ikeda sum rule which is reduced compared to the value $3(N-Z)$
and results equal to $3(N-Z)$ times a quenching factor $q^2 \sim 0.84 - 0.92$. The authors also checked that the value $3(N-Z)$ is obtained if the transition operator contains only a one-body part. Hence, they concluded that the inclusion of two-body currents may solve the
discrepancy between theoretical GT integrated strengths and experimental measurements.
They found that 70-80 \% of the total (reduced) strength was exhausted for the three nuclei, up to the excitation energy of 10 MeV. Similar conclusions were drawn in Ref. \cite{gysbers}, where
the $\beta$-decay rate was computed for $^{100}$Sn within the same theoretical framework.
We show the results of Ref. \cite{ekstrom} in Fig. \ref{ik}, together with RPA and SSRPA predictions obtained with SGII, for the nuclei $^{14}$C and $^{22}$O, as illustrations.
As for the case of $^{48}$Ca, we have checked that also for these lighter nuclei a hybrid calculation (diagonal approximation in the matrix to invert) reveals itself to be a rather poor approximation and a full calculation is necessary.
The RPA and SSRPA cumulative sums for the $S_{GT^-}$ strength are plotted in the interval of excitation energies from 4 to 30 MeV. The excitation energy of 10 MeV is indicated by a vertical dashed magenta line. The $S_{GT^-}$ strength is dominant compared to the $S_{GT^+}$ one, as one can easily check in the figure, where the RPA cumulative sums almost attain the total Ikeda values of 18 and 6 for
$^{22}$O and $^{14}$C, respectively. The two horizontal colored areas represent, for the two nuclei, the total Ikeda sum rule reduced by the $q^2$ value deduced in Ref. \cite{ekstrom}. The vertical blue and green intervals at 10 MeV correspond to 70-80 \% of each of these reduced sum rules. We may notice that, at 10 MeV, the RPA cumulative sums are already above these vertical intervals in both cases, whereas the SSRPA values are located below.
The SSRPA values remain below the two colored horizontal areas in the whole interval of energies shown in the figure, up to 30 MeV.
This indicates that the way the quenching is accounted for in our model (where the total sum rule $3(N-Z)$ is fulfilled) provides coherent results compared to those of Ref. \cite{ekstrom}, contrary to what happens in RPA, where the values are located above the two horizontal areas already at 8 MeV of excitation energy.
This also means that, in practice, the inclusion of a two-body part in the transition operator does not seem to be an important ingredient for the description of the quenching in our model. We finally stress that we prefer to work within a model where the total sum rule $3(N-Z)$ is expected to be exhausted (even if at high excitation energies), because this provides a precise reference value and, thus, a robust check for the theoretical calculations.
\begin{figure}
\includegraphics[scale=0.32]{ikedatest.eps}
\caption{RPA and SSRPA cumulative sums for the $S_{GT^-}$ strength for the nuclei $^{22}$O (blue) and $^{14}$C (green). The blue and green horizontal areas represent the reduction of the total Ikeda sum rule $S_{GT^-}-S_{GT^+}$ that was predicted for these nuclei in Ref. \cite{ekstrom}. The vertical dashed magenta line indicates 10 MeV of excitation energy.
The blue and green vertical intervals there correspond to the predictions of Ref. \cite{ekstrom} for the amount (70-80 \%) of the total (reduced) sum rule exhausted up to 10 MeV for the two nuclei (see text). }
\label{ik}
\end{figure}
\section{Heavier nuclei: $^{90}$Zr and $^{132}$Sn} \label{heavy}
The $^{90}$Zr(p,n) and $^{90}$Zr(n,p) reactions were performed at the Research Center for Nuclear Physics and the results were published in 1997 \cite{zr1}
and in 2005 \cite{zr2}, respectively. A consistent study of data coming from both (p,n) and (n,p) channels led in Ref. \cite{zr2} to the estimation of the experimental quenching for the GT Ikeda sum rule. In particular, the GT$^-$ strength integrated up to 50 MeV was found to be equal to 29.3, very close to the full value of the Ikeda sum rule for this nucleus ($3(N-Z)=$ 30), whereas the GT$^+$ strength integrated up to the same excitation energy was found to be equal to 2.9.
\begin{figure}
\includegraphics[scale=0.32]{dfff-Zr.eps}
\caption{GT$^-$ strengths obtained with the Skyrme interaction SGII for the nucleus $^{90}$Zr in MeV$^{-1}$. SSRPAFF and SSRPADF spectra are shown. }
\label{diago}
\end{figure}
We show in Fig. \ref{diago} the comparison between the SSRPADF and SSRPAFF GT$^-$ spectra in $^{90}$Zr, having folded the discrete spectra with a 1-MeV-width Lorentzian. Differently from the case of lighter nuclei, the hybrid approximation, where the matrix to invert is treated as diagonal, seems to work very well in this case. The two spectra are indeed very similar and almost superposed, especially at higher energies. We may conclude that, when the number of 2p2h configurations becomes very large in nuclei located beyond the medium-mass region (the number of configurations becomes typically larger that $5 \cdot 10^4$) the off-diagonal terms may be safely neglected in the matrix to invert without loosing any important information in the excitation spectrum. We have indeed checked that this is the case by using different kind of Skyrme forces. In the present section, we will thus present SSRPADF calculations for the two heavier nuclei that we treat.
To see how the fine structure of the excitation spectrum is described, we show in Fig. \ref{fineZr} the SSRPA and RPA GT$^{-}$ discrete spectra (no units) and the experimental spectrum of Ref. \cite{zr2} (in MeV$^{-1}$).
The theoretical spectra are not folded to avoid any widths and fragmentations artificially induced by the folding itself, especially in the RPA case, where the number of discrete peaks is very low.
The absolute values of the theoretical bars and the experimental strengths cannot of course be compared to each other since they are expressed in different units, whereas it is meaningful to compare the positions of the peaks and the fragmentation of the strength.
To better visualize all the results in the same plot, the discrete spectra have been scaled, the RPA one being divided by 7.3 and the SSRPA one being multiplied by 5.
We immediately observe that, whereas the RPA spectrum displays only two discrete main peaks, the SSRPA one has a quite dense distribution of peaks, describing much more realistically the experimental distribution of the strength. We see that, with the used interaction SGII, the SSRPA spectrum is slightly shifted to lower energies compared to the experimental one whereas the RPA one slightly overestimates the excitation energies.
\begin{figure}
\includegraphics[scale=0.32]{fine-Zr.eps}
\caption{GT$^-$ SSRPA and RPA discrete spectra obtained with the Skyrme interaction SGII for the nucleus $^{90}$Zr (no units). The RPA discrete spectrum is divided by 7.3 and the SSRPA one is multiplied by 5.
The experimental results, in MeV$^{-1}$, are extracted from Ref. \cite{zr2}. }
\label{fineZr}
\end{figure}
The same analysis can be done for the heavier nucleus $^{132}$Sn. The experimental (p,n) reaction was carried out for this nucleus at the Radioactive Isotope Beam Factory in RIKEN and the results were published in Ref. \cite{sn}. Also for this nucleus we compare, in Fig. \ref{fineSn}, the discrete RPA and SSRPA spectra (no units) with the corresponding experimental distibution (in MeV$^{-1}$).
Again, since the absolute values of the theoretical and experimental strengths are not comparable, the RPA discrete spectrum has been divided by 7 and the SSRPA one has been multiplied by 14 to better visualize all the results in the same plot.
The RPA spectrum has much less discrete peaks than the SSRPA one which provides naturally a fragmented and more dense response distribution. Again, the SSRPA spectrum is slightly shifted towards lower energies compared to the experimental one whereas the RPA one slightly overestimates the excitation energies.
\begin{figure}
\includegraphics[scale=0.32]{fine-Sn.eps}
\caption{GT$^-$ SSRPA and RPA discrete spectra obtained with the Skyrme interaction SGII for the nucleus $^{132}$Sn (no units). The RPA spectrum is divided by 7 and the SSRPA one is multiplied by 14. The experimental results, in MeV$^{-1}$, are extracted from Ref. \cite{sn}. }
\label{fineSn}
\end{figure}
The fact that we have found, for the two nuclei, an overall small shift of the spectra to lower energies compared to experimental results is something that depends on the choice of the effective interaction. We may in fact expect that other Skyrme interactions would provide spectra slightly shifted towards higher or lower energies. For example, it is known that a class of Skyrme interactions, called SAMi, was introduced in Ref. \cite{roca} and tailored specifically for describing spin-isospin excitations within the RPA model (and only within this model).
Whereas the fragmentation will be described in a similar way by many different Skyrme interactions (this represents a genuine SSRPA effect), the specific position of the peaks in the spectrum will be slightly interaction-dependent.
As a matter of fact, another genuine SSRPA effect can be identified in the integrated strengths and, in particular, in their quenching.
In this respect, we show in Fig. \ref{quen} the folded SSRPA and RPA strengths (the folding is done with a 1-MeV-width Lorentzian) and the experimental strengths for $^{90}$Zr (a) and $^{132}$Sn (c), together with the cumulative sums of the strengths, evaluated up to 25 MeV for $^{90}$Zr (b) and up to 20 MeV for $^{132}$Sn (d).
\begin{figure}
\includegraphics[scale=0.32]{cumu.eps}
\caption{(a) GT$^-$ SSRPA and RPA folded spectra obtained with the Skyrme interaction SGII for the nucleus $^{90}$Zr.
The experimental results Ref. \cite{zr2} are also shown; (b) Strenghts of panel (a) integrated up to 25 MeV; (c) Same as in (a) but for the nucleus $^{132}$Sn. The experimental results are extracted from Ref. \cite{sn}; (d) Strenghts of panel (c) integrated up to 20 MeV. }
\label{quen}
\end{figure}
For the nucleus $^{90}$Zr, calculations carried out by Drozdz et al. \cite{dro} (within the SRPA model) and by Bertsch and Hamamoto \cite{bertsch} (within a perturbative approach) were reported in Ref. \cite{zr1}. However, those old SRPA calculations could be carried out, at that time, only by resorting to strong cuts and approximations. It is known that, during the last years, this theoretical model was strongly improved and refined both in its formal aspects and in its numerical implementations, for example by ourselves \cite{gamba2020,gamba2015}.
We thus cannot regard anymore those old results as robust references to which compare our present predictions.
A modern beyond-mean-field computation for the GT response in $^{90}$Zr was recently reported in Ref. \cite{robin}. These calculations were carried out within an approach based on the relativistic particle-vibration coupling and, for the first time, ground state correlations were included in the parent nucleus, coming from the particle-vibration coupling itself. A good agreement with the experimental strength distribution was found by the authors and commented in the article. However, the cumulative sum of the strength was not computed and shown there (to which we could compare our integrated strength). The authors of Ref. \cite{robin} mentioned by the way in this respect that the inclusion of the new correlations in the ground state of the parent nucleus induces a small violation of the Ikeda sum rule, as was already discussed in Ref. \cite{tse2007}.
The only integrated strength we can refer to for this nucleus is the one coming from RPA calculations, which is shown in Fig. \ref{quen} for SGII. Other RPA strength distributions and cumulative sums were also shown for $^{90}$Zr for example in Ref. \cite{cao}, obtained by using different Skyrme interactions. However, independently of the interaction, the RPA strength integrated up to 25 MeV is in all cases already almost equal to 30, much larger compared to the experimental value (b). Panel (b) shows that the SSRPA computation leads to a strongly improved agreement with the experimental integrated strength, providing a much lower value than in RPA (22.3) at the excitation energy of 25 MeV. The corresponding experimental value is slightly above 20, as one can see in the figure.
The slope of the SSRPA curve nicely follows the experimental one and the sligth shift which is visible is generated by the already mentioned small shift (to lower values) in the excitation energies compared to the experiment.
For the nucleus $^{132}$Sn, modern beyond-mean-field particle-vibration-coupling calculations are available in the literature, within both non-relativistic \cite{niu} and relativistic \cite{robin2} approaches. However, only the authors of Ref. \cite{niu} presented in their article integrated strengths up to 20 MeV of excitation energy .
Their results depend on the choice of a smearing parameter $\Delta$. By increasing it, they can reduce the value of the cumulative sum. With the largest used value, $\Delta=1$ MeV, they obtain their best result, that is an integrated strength of 80 (at 20 MeV). Even if our prediction at 20 MeV is located above the experimental value (we notice that the overall shift of the SSRPA spectrum to lower energies is also responsible for this) we may observe that we obtain anyway a better result than the one of Ref. \cite{niu}, 10 units lower than the particle-vibration coupling prediction.
We may thus conclude that, in general, the CE-SSRPA model seems to be more effective than other available beyond-mean-field EDF approaches in describing the quenching of GT strengths, globally leading to a much better agreement with the experimental measurements.
\section{Conclusions} \label{conclu}
We have applied the CE-SSRPA model, recently introduced in Ref. \cite{gamba2020} for $^{48}$Ca and $^{78}$Ni, to other closed-shell and closed-subshell nuclei located in different regions of the nuclear chart.
A preliminary analysis is carried out in $^{48}$Ca to check the validity of different approximations. This study reveals that, for this nucleus (and for the lighter ones that we have treated here) the inversion of the $A_{22}$ matrix in the subtraction procedure must be performed without any type of approximations. If in particular the matrix to invert is approximated as diagonal the quality of the predictions results deteriorated, although a considerable improvement with respect to RPA is still obtained.
Two lighter nuclei are analyzed, $^{14}$C and $^{22}$O to compare the SSRPA cumulative sums of the strength with the {\it{ab-initio}} coupled-cluster results of Ref. \cite{ekstrom}. This comparison indicates that, differently from the case of Ref. \cite{ekstrom}, the inclusion of a two-body part in the GT transition operator does not seem to be a crucial ingredient within our model to account for the quenching of the GT Ikeda sum rule.
Finally, two heavier nuclei are investigated, $^{90}$Zr and $^{132}$Sn. We have checked that, for these heavier nuclei, where the number of 2p2h configurations considerably increases, the matrix to invert in the subtraction procedure can be simplified and safely taken as diagonal, without inducing important changes in the spectra.
The analysis of the strengths integrated up to excitation energies between 20 and 25 MeV reveals that, as was already found for $^{48}$Ca, the CE-SSRPA model incorporates the necessary correlations to provide a much more effective description of the quenching of GT spectra and, consequently, a much better agreement with experimental data.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 4,453
|
\section{\bf Introduction}
The memory effect in gravitational wave (GW) signals is the phenomenon, wherein the relative position of two inertial GW detectors undergoes a permanent displacement owing to the passage of GWs through them \cite{Zeldovich-1, Grishchuk:1989qa, Thorne:1991}. The memory effect is categorised broadly according to the nature of the source of GW. Linear memory arises from sources which have unbound components, like hyperbolic binary orbits \cite{Turner-1, Turner-Will}, neutrinos from supernova \cite{Epstein:1978dv, Turner-SN, Kotake:2005zn, Andresen:2016pdt, Morozova:2018glm, Vartanyan:2020nmt, Mukhopadhyay:2021zbt}, gamma-ray bursts \cite{Sago:2004pn,Urrutia:2022lce,Piran:2022gxr} and exotic objects like cosmic strings \cite{Jenkins:2021kcj}.
Another type of memory signal is the non-linear memory \cite{Christodoulou:1991,Will:1991, Blanchet:1992br, Thorne:1992sdb}, where secondary gravitational waves are produced by the primary gravitational waves from sources like coalescing binaries. The importance of non-linear memory is that it will be an experimental proof of graviton-graviton coupling or the non-linear nature of the gravitational field equations in Einsteinian gravity.
The memory signal has a significance in field theory as it follows from the soft-graviton theorems \cite{Weinberg-0, Weinberg-1, Weinberg-2} where the amplitude of a low energy graviton emission from a scattering process can be related, by a multiplicative kinematic factor, to the hard scattering amplitude without the graviton emission. The zero-graviton frequency amplitude of the soft-amplitude has a pole in the frequency space which in Fourier space is a step function in time which is the characteristic of the memory signal. Weinberg's soft theorem amplitude has been generalised to include higher order terms in graviton momenta using the gravitational gauge invariance an angular momentum conservation \cite{Cachazo:2014fwa}. Calculations of graviton emission amplitudes show that there are non-analytic logarithmic terms in graviton frequencies even in tree level scattering \cite{Laddha:2018myi,Sahoo:2018lxl,Laddha:2018vbn}.
The non-linear memory signal, for binaries in a quasi-circular orbit, already occurs at the 0-PN order \cite{Will:1991, Kenneflick-94}. In \cite{Favata:2008yd, Favata:2010zu} the non-linear memory for the quasi-circular orbit was computed at the 3-PN (Post-Newtonian) order. The non-linear memory signal from eccentric binary orbits was calculated at 3-PN order in \cite{Favata:2011qi} and the 3-PN calculation for eccentric orbits including the tail contributions is done in \cite{Ebersold:2019kdc}.
Binaries (like the Hulse-Taylor) can have large initial eccentricities but by the time their frequencies enter the threshold of detectors like Advanced LIGO with a threshold of $\sim 10$ Hz, they lose their eccentricities due to gravitational radiation reaction. However, there can be other initial configurations of binary star orbits which could give rise to large eccentricity orbits by the time the frequency enters the Advanced LIGO threshold \cite{Samsing:2017xmd, Tucker:2021mvo}. An eccentric binary can be formed by capture from an unbound orbit \cite{Quinlan, OLeary:2008myb}. An initial three body system may eject one of the bodies and result in an eccentric binary at coalescence \cite{Heggie}. A three body system in which the orbit of one body is at a larger radius, may cause the distant body to perturb the orbit of the tight binary system. This perturbation may give rise to Kozai-Lidov oscillations \cite{Kozai,Lidov} in the binary orbit which can drive the eccentricity to large values \cite{Wen:2002km, Naoz:2012bx}. Constructing gravitational wave templates for coalescent eccentric binaries is therefore of importance for observations \cite{Tucker:2021mvo} at future detectors like Advanced-LIGO. A determination of eccentricity of the orbits of the coalescent binary will be possible at Advanced-LIGO by improving the low frequency sensitivity \cite{Favata:2021vhw}. Experimental determination of both the memory signal and the eccentricity of the orbit affect the signal near the low-frequency threshold of the detector, therefore construction of memory signal templates must accurately take into account the eccentricity of the orbit.
The non-linear gravitational wave memory from binary mergers may be discernible through next generation ground based detectors like Advanced LIGO \cite{LIGOScientific:2014pky} and Advanced Virgo \cite{VIRGO:2014yos} by improving the sensitivity in the 5-10 Hz band \cite{Johnson:2018xly, Yang:2018ceq,Yu:2017zgi, Talbot:2018sgr}. The memory effect for individual binaries may be resolvable by forthcoming interferometers such as LISA \cite{LISA:2022kgy}, Cosmic Explorer \cite{LIGOScientific:2016wof} and Einstein Telescope \cite{Einstein-Telescope}. Gravitational-wave memory from binary mergers may also be seen in Advanced LIGO by combining the signals from multiple events \cite{Lasky:2016knh}. The memory signal from several unresolved binary event mergers may be observed as a cumulative change over time in the pulsar timing residuals and may be observed in pulsar timing arrays (PTAs) \cite{ Maiorano:2021sqj, NANOGrav:2019vto, Burke-Spolaor:2018bvk, Rosado:2015epa, Pshirkov:2009ak}. Ground-based detectors are sensitive to the frequency of the GW in the $5-10^3$ Hz band. Space-based detector LISA will probe frequencies as low as $10^{-3}$ Hz,
and pulsar timing arrays may measure frequency as low as $10^{-9}$ Hz.
Black-holes in hyperbolic orbits in the gravitational field of super-massive black-holes at the galactic center radiate bursts of gravitational waves \cite{ Capozziello:2008ra, DeVittori:2012da,DeVittori:2014psa,Garcia-Bellido:2017knh, Garcia-Bellido:2017qal,Grobner:2020fnb, Cho:2018upo} which may be detectable at Advanced LIGO or Einstein Telescope \cite{Mukherjee:2020hnm, Codazzo:2022aqj}.
For hyperbolic orbits the nonlinear memory appears at the 2.5-PN level in the waveforem. This is contrary to the case of elliptical and quasicircular orbits, where the nonlinear memory appears at the Newtonian (0-PN) order \cite{Will:1991, Favata:2011qi}. This is because the radiation reaction effects accumulate over time in the closed orbits while for open orbits the radiation reaction is maximum at the closest approach and is zero at asymptotic past and future times.
The Baysian inference of specific signals from the data is most efficiently achieved by match filtering of signals in the frequency domain with the data \cite{Khan:2015jqa,Garcia-Quiros:2020qpx}. With this aim, we compute the wave-forms for the memory signals for eccentric elliptical and hyperbolic orbits in the frequency domain. We compute the waveforms in using the tree level graviton emission amplitude with the stress tensor of the binary orbits as sources. Using this graviton emission amplitude we compute the frequency spectrum of the energy radiated following \cite{Mohanty:1994yi,KumarPoddar:2019jxe,KumarPoddar:2019ceq,KumarPoddar:2019ceq,Poddar:2021yjd}. For elliptical orbits, the energy radiation spectrum is the source of the non-linear gravitational memory which we thus obtain directly in the frequency space.
The anatomy of this article can be described as follows: In section \ref{sec:QFT-appr}, we outline how the gravitational waveform can be constructed in a field-theoretic approach and we also highlight the computation of the rate of energy loss associated with gravitational wave radiation. In section \ref{sec:hyperbolic}, we compute components of the stress tensor, in frequency domain, for binaries in hyperbolic orbits. This is followed by the evaluation of the same quantities in the limit of vanishing frequency and the associated linear memory in section \ref{sec:linear-memory}. Next, we elucidate the general formalism for the non-linear memory associated with the radiation from binaries in closed orbits in section \ref{sec:nonlinear-memory}. We delve into the details for the special case of elliptical orbits in section \ref{sec:elliptical}, where we compute the transverse-traceless memory waveform. We summarize our conclusions in section \ref{sec:conclusion}.
\section{\bf Gravitational waveform and energy radiated from scattering amplitudes}\label{sec:QFT-appr}
The probability amplitude of emitting a graviton of polarisation $\epsilon^\lambda_{\mu \nu}(\vec n)$ from a source with stress tensor (in the momentum space ) ${\widetilde T}^{\mu \nu} (k)$ is given by
\begin{eqnarray}\label{amp-1}
{\cal A}_\lambda(k_0, \vec n k_0)
=-\iota\,\cfrac{\kappa}{2}\, \epsilon_{\mu \nu} ^{*\lambda}(\vec n)\,
{\widetilde T}^{\mu \nu}(k_0, \vec n \, k_0 ) \,.
\end{eqnarray}
We can express the gravitational wave metric observed at the detector in terms of the probability amplitude of a graviton emission by a source at a distance $r$ as,
\begin{eqnarray} \label{Greens-3}
h_{\alpha \beta}(\vec x, t)= \cfrac{1}{4 \pi r} \, \int \, \cfrac{d k_0}{(2 \pi)}\, \sum_{\lambda=1}^2\, \epsilon_{\alpha \beta} ^{\lambda}(\vec n)\, {\cal A}_\lambda(k_0, \vec n\, k_0)\, e^{-\iota k_0 (t-r) }\,.
\end{eqnarray}
The graviton field in Eq.~\eqref{Greens-3} is a canonical spin-2 field with mass dimension 1 as it is defined as an expansion of the metric $g_{\mu \nu}=\eta_{\mu \nu} + \kappa\, h_{\mu \nu}$ where $\kappa=\sqrt{32 \pi G}$. The metric perturbation identified as gravitational wave is the dimensionless quantity $\widetilde h_{\mu \nu}\equiv g_{\mu \nu}- \eta_{\mu \nu}= \kappa\, h_{\mu \nu}$. The expression for the dimensionless gravitational wave in terms of the amplitude is therefore from Eq.~\eqref{Greens-3} given by
\begin{eqnarray}\label{Greens-4}
\widetilde h_{\alpha \beta}(\vec x, t)= \cfrac{\kappa}{4 \pi r} \,\int\, \cfrac{d k_0}{(2 \pi)} \, \sum_{\lambda=1}^2 \, \epsilon_{\alpha \beta} ^{\lambda}(\vec n) \, {\cal A}_\lambda(k_0, \vec n\, k_0)\, e^{-\iota k_0 (t-r) } \,.
\end{eqnarray}
This relates the waveform at the detector to the probability amplitude of graviton emission by the source. To relate the waveform at the detector to the source stress tensor we substitute in Eq.~\eqref{Greens-4}, the expression for the amplitude given in Eq.~\eqref{amp-1}, to obtain
\begin{eqnarray}\label{Greens-5}
\widetilde h_{\alpha \beta}(\vec x, t)&=& -\cfrac{\kappa^2}{8 \pi r} \, \int \, \cfrac{dk_0}{2\pi} \, \sum_{\lambda=1}^2 \, \epsilon_{\alpha \beta} ^{\lambda}(\vec n) \, \epsilon_{\mu \nu} ^{*\lambda}(\vec n)\,
{\widetilde T}^{\mu \nu}(k_0, \vec n\, k_0 )\, e^{-\iota k_0 (t-r) }\,\nonumber\\
&=&-\cfrac{4G}{ r} \,\int \,\cfrac{dk_0}{2\pi} \,\left(\widetilde T_{\alpha \beta}(k_0, \vec n\, k_0 ) - \cfrac{1}{2}\, \eta_{\alpha \beta}\, {\widetilde T^\mu}_\mu (k_0, \vec n\, k_0 ) \right) \, e^{-\iota k_0 (t-r) }\,.
\end{eqnarray}
where we made use of the completeness relation
\begin{eqnarray}\label{Polsum-mu}
\sum_{\lambda=1}^2\, \epsilon_{\mu\nu}^\lambda(k)\,\epsilon_{ \alpha\beta}^{*\lambda}(k) = \cfrac{1}{2}\,(\eta_{\mu\alpha}\,\eta_{\nu\beta}+\eta_{\mu\beta}\,\eta_{\nu\alpha})- \cfrac{1}{2}\,\eta_{\mu\nu}\,\eta_{\alpha\beta}.
\end{eqnarray}
To obtain the propagating degrees of freedom, we need to project the transverse traceless (TT) components of the wavefunction constructed in Eq.~\eqref{Greens-5}.
\begin{eqnarray}
\Big[\widetilde h_{ij}\Big]^{\text{TT}}(\vec x, t) = -\cfrac{4G}{ r}\,\Lambda_{ij,kl}(\vec n) \, \int \, \cfrac{dk_0}{2\pi} \, \left(\widetilde T_{k l}(k_0, \vec n\, k_0 ) - \cfrac{1}{2}\, \eta_{k l}\, { \widetilde T^\mu}_{\,\,\,\,\mu} (k_0, \vec n\, k_0 ) \right)\, e^{-\iota k_0 (t-r) }.
\end{eqnarray}
where, $ \Lambda_{ij,kl}(\vec n)$ is the transverse-traceless projection operator defined with respect to the direction, $\hat n$, of the emitted gravitational wave. The explicit form of the TT projection operator is
\begin{eqnarray}\label{tensor-1}
\Lambda_{ij,kl}( \hat n)&=&P_{ik}( \hat n) P_{jl} ( \hat n)- \cfrac{1}{2} P_{ij}( \hat n)P_{kl} ( \hat n) = \left(\delta_{ik}- n_i n_k\right)\left(\delta_{jl} - n_j n_l\right)-\cfrac{1}{2} \left(\delta_{ij}- n_i n_j \right)\left( \delta_{kl}- n_k n_l\right).
\end{eqnarray}
Since $\Lambda_{ij,kl} \,\,\eta_{k l}=0$, the ${\widetilde T^\mu}_{\,\,\,\,\mu}$ term vanishes and we obtain the simpler result,
\begin{eqnarray}\label{GWTmunu}
\Big[\widetilde h_{ij}\Big]^{\text{TT}}(\vec x, t)= -\cfrac{4G}{ r}\,\Lambda_{ij,kl}(\vec n)\, \int \, \cfrac{dk_0}{2\pi}\, T_{k l}(k_0, \vec n\, k_0 ) \, e^{-\iota k_0 (t-r) }.
\end{eqnarray}
In frequency space, the observed gravitational wave and the stress-tensor of the source can be related as,
\begin{eqnarray}\label{GWTmunu-f}
\Big[\widetilde h_{ij}\Big]^{\text{TT}}(\vec x, k_0)= -\cfrac{4G}{ r}\, \Lambda_{ij,kl}(\vec n)\, T_{k l}(k_0, \vec n\, k_0 ).
\end{eqnarray}
We shall use Eq.~\eqref{GWTmunu-f} to compute the gravitational waveform from various sources, like compact binaries in bound and unbound orbits, by computing the stress-tensor of the source in frequency space.
\subsection{\bf Power spectrum of gravitational wave in field theoretic approach}
The rate of graviton emission is given by the Fermi Golden Rule, and is the amplitude squared summed over the final state graviton polarisation and integrated over the phase space volume,
\begin{eqnarray}\label{dgamma0}
\Gamma &=&\sum_\lambda\, \int\, \cfrac{|S_{fi}|^2}{T} \,
\cfrac{d^3 \vec k}{(2 \pi)^3\, 2\omega} = \sum_n \,\sum_\lambda \, \int\, \big\vert {\cal A}_\lambda(\omega, \omega^\prime_n) \big\vert^2 \,(2 \pi)\, \delta(\omega-\omega_n^\prime)\,
\cfrac{d^3 \vec k}{(2 \pi)^3\, 2 \omega} \nonumber\\
&=&\cfrac{ \kappa^2}{4}\, \sum_n \, \sum_\lambda \, \int \, \big\vert T_{\mu \nu}( \vec k, \omega_n^\prime) \, \epsilon_\lambda^{*\mu \nu} (\vec k) \big\vert^2 \,(2 \pi)\, \delta(\omega-\omega_n^\prime)\,
\cfrac{d^3 \vec k}{(2 \pi)^3\, 2\omega_k}.
\end{eqnarray}
Here, $\lambda$ accounts for the possible states of polarization and $n$ denotes the harmonics corresponding to the emission. The rate of energy radiated is obtained from the probability of radiation given above by including an extra factor of $\omega_k=|\vec k|$ in the integral,
\begin{eqnarray}\label{dEdt-1}
\cfrac{dE_\text{gw}}{dt}&=&\cfrac{ \kappa^2}{4}\, \sum_n \sum_\lambda \, \int \,\big\vert T_{\mu \nu}( \vec k, \omega_n^\prime)\, \epsilon_\lambda^{*\mu \nu} (\vec k) \big\vert^2 \,(2 \pi)\, \delta(\omega-\omega_n^\prime)\, \,\, \omega_k \,\,\cfrac{d^3 \vec k}{(2 \pi)^3\, 2 \omega}.
\end{eqnarray}
We can open the modulus squared piece of the integrand in Eq.~\eqref{dEdt-1} and simplify using the polarisation sum relation, see Eq.~\eqref{Polsum-mu}, as follows:
\begin{eqnarray}\label{Tmunusquare}
\sum_\lambda \, \big\vert T_{\mu \nu}( \vec k, \omega_n^\prime)\, \epsilon_\lambda^{*\mu \nu} (\vec k) \big\vert^2 &=& \sum_\lambda \, \left( T_{\mu \nu}( \vec k, \omega_n^\prime)\, T_{\alpha \beta}^*( \vec k, \omega_n^\prime)\right)\left(\epsilon_\lambda^{*\mu \nu} (\vec k) \, \epsilon_\lambda^{\alpha \beta} (\vec k) \right)\nonumber\\
&=&T_{\mu \nu}( \vec k, \omega_n^\prime)\, T^{*\nu \mu}( \vec k, \omega_n^\prime) -\cfrac{1}{2}\, \big\vert {T^\mu}_{\mu}( \vec k, \omega_n^\prime) \big\vert^2\,.
\end{eqnarray}
The $T^{00}$ and $T^{i0}$ components can be expressed in terms of the $T^{ij}$ ones by utilizing
the conserved current relation, $k_\mu T^{\mu\nu}=0$. This allows us to write
\begin{equation}\label{app6}
T_{0j}=-\hat{k^i}\,T_{ij},\hspace{0.5cm} T_{00}=\hat{k^i}\,\hat{k^j}\,T_{ij}.
\end{equation}
Using these relations, we can rewrite Eq.~\eqref{Tmunusquare} as
\begin{eqnarray}\label{app7}
\big\vert T_{\mu\nu}(\vec k, \omega_n^\prime)\big\vert^2 -\cfrac{1}{2}\, \big\vert T^{\mu}{}_{\mu}(\vec k, \omega_n^\prime)\big\vert^2 &=& T_{ij}T^{*ji}+T_{00}T^{*00}+ T_{0i}T^{*i0}+ T_{i0}T^{*0i} - \cfrac{1}{2}\left({T^0}_0+{T^i}_i \right)\left({T^{*0}}_0+{T^{*j}}_j \right)\nonumber\\
&=&\left(T_{ij}T^{*ji} -\cfrac{1}{2}\, T^i_i T^{*j}_j\right) +\cfrac{1}{2} \,\hat{k^i}\hat{k^j}\hat{k^l}\hat{k^m} T_{ij} T^*_{lm} - \left( \hat{k^l}\,\hat{k^m}\, T_{il} T^*_{mi} + \hat{k^l}\,\hat{k^m}\, T_{il} T^*_{mi}\right)
\nonumber\\
&& \qquad\,+\, \cfrac{1}{2}\,\left(\hat{k^l}\,\hat{k^m} \,T^*_{lm} T^{i}_i + \hat{k^l}\,\hat{k^m} \, T_{lm} T^{*j}_j \right).
\end{eqnarray}
In the quadrupole approximation of the source, for sources smaller in size than the wavelength of the GWs, $\vec k \cdot \vec x \ll 1$, the stress tensor in momentum space $T_{\mu\nu}( \vec k, \omega_n^\prime)$ has no explicit $\vec{k}$ dependence. Therefore, after substituting the contents of Eq.~\eqref{app7} in Eq.~\eqref{dEdt-1}, one can perform the angular integrations using the following relations:
\begin{eqnarray}\label{dOmegak}
&&\int d\Omega_k =4 \pi, \hspace{0.9cm}\int d\Omega_k \,\hat{k^i}\,\hat{k^j}=\cfrac{4\pi}{3}\,\delta_{ij}, \hspace{0.7cm} \int d\Omega_k \, \hat{k^i}\,\hat{k^j}\,\hat{k^l}\,\hat{k^m}=\cfrac{4\pi}{15}\,\left(\delta_{ij}\,\delta_{lm}+\delta_{il}\,\delta_{jm}+\delta_{im}\,\delta_{jl} \right).
\end{eqnarray}
to obtain
\begin{equation}\label{app9}
\int\, d\Omega_k \, \Big[|T_{\mu\nu}(\vec k, \omega_n^\prime)|^2-\cfrac{1}{2}\,|T^{\mu}{}_{\mu}(\vec k, \omega_n^\prime)|^2\Big]
=\cfrac{8\pi}{5}\,\Big(T_{ij}(\omega_n^\prime)\,T^*_{ji}(\omega_n^\prime)-\cfrac{1}{3}\,|T^{i}{}_{i}(\omega_n^\prime)|^2\Big),
\end{equation}
Finally, using the result of Eq.~\eqref{app9}, the expression for the energy radiated by a source in terms of the source stress tensor can be obtained as a modification of Eq.~\eqref{dEdt-1} as
\begin{eqnarray}\label{dEdt-2}
\cfrac{dE_\text{gw}}{dt}=\cfrac{ \kappa^2}{4} \, \sum_n\, \int \, \cfrac{8 \pi}{5}\,\Big(T_{ij}(\omega_n^\prime)T^*_{ji}(\omega_n^\prime)-\cfrac{1}{3}\,|T^{i}{}_{i}(\omega_n^\prime)|^2\Big)\,{\omega}^2\, 2\pi \,\delta(\omega_n^\prime-\omega) \cfrac{d\omega}{(2 \pi)^3\, 2 \omega} .
\end{eqnarray}
We make use of this expression for computing the energy radiated by binaries in elliptical orbits which is the source of the non-linear memory signal. We will also show using this expression that in the memory signal from hyperbolic orbits, there is non-zero energy radiated around the zero-frequency band.
\section{\bf Hyperbolic Orbits}\label{sec:hyperbolic}
For the case of unbounded orbits, we consider a black-hole of mass $m_2$ in a hyperbolic orbit around a larger black-hole of mass $m_1$. Similar to the case of the elliptical orbits, in the centre of mass frame, the motion is equivalently described by a single body having the reduced mass $\mu = \cfrac{m_1\,m_2}{m_1+m_2}$. The following quantities describe the system:
\begin{itemize}
\item The coordinates parametrized as,
\begin{eqnarray}
x(\xi) = a\,(e - \cosh\xi),\qquad y(\xi) = b\,\sinh\xi, \qquad z(\xi) = 0, \qquad \cfrac{\omega^\prime}{\nu}\,t = \Omega\, t = (e\sinh\xi - \xi),
\end{eqnarray}
where, $a$ and $b \,\,(= a \sqrt{e^2 - 1},\,\, e > 1)$ denote the semi-major and semi-minor axes and the variables $e$ and $\xi \in (-\infty, \infty)$ refer to the eccentricity and the hyperbolic anomaly of the orbit respectively.
\item The angular frequency $\omega^\prime$ is proportional to the fundamental frequency $\Omega$, as $\omega^\prime = \nu \,\Omega$, where $\nu \in (0, \infty)$ is a non negative real number and $\Omega = \left(\cfrac{G\, (m_1 + m_2)}{a^3}\right)^{1/2}$.
\end{itemize}
We start with the computation of the stress-tensor components in frequency space. Focussing first on the $xx$ component, the calculation proceeds as follows:
\begin{eqnarray}
T_{xx}(\omega^{\prime})&=& \int_{-\infty}^{\infty} dt \,e^{i \omega^\prime t} \,\mu \,\dot x^2 \,= \, -\iota \, \mu \, \omega^\prime \int_{-\infty}^{\infty} dt \,e^{i \omega^\prime t} \, \,x\, \dot x \,= \, -\iota \,\mu \,\omega^\prime \int_{-\infty}^{\infty} d\xi \,e^{i \nu(e \sinh \xi -\xi)} \, \,x\, \cfrac{ dx}{d\xi},
\end{eqnarray}
where, the second equality is obtained after implementing integration by parts and neglecting a term proportional to $\ddot{x}(t)$. The final expression is obtained after incorporating a change of variables $t \rightarrow \xi$. Since, the integration is over $\xi$, the result will simply be a function of $\nu$. Also, since $\omega^\prime = \nu \Omega$, it makes more sense to express $T_{ij}$ as a function of $\nu$ instead of $\omega^\prime$. Thus, we obtain
\begin{eqnarray}\label{eq:Txx-hyperb}
T_{xx}(\nu) &=& \, \iota \,\mu \, \nu\, \Omega \, a^2 \int_{-\infty}^{\infty} d\xi \,e^{\iota \nu(e \sinh \xi -\xi)} \, \, \sinh \xi\,\, (e- \cosh\xi) \, \nonumber\\
&=& \mu\,\nu\, \Omega\, a^2 \pi\, \left[ \, \cfrac{\iota}{\nu e^2}\, H^{(1)}_{\iota\nu}(ie\nu) - \left(e - \cfrac{1}{e}\right){H_{\iota\nu}^{(1)}}^\prime(\iota e\nu)\, \right].
\end{eqnarray}
We replaced $\sinh \xi$, $\cosh \xi$ by their exponential counterparts. Subsequently, we identified and replaced the integrals with Hankel functions, i.e.,
\begin{eqnarray}\label{eq:Hankel-1}
H_p^{(1)}(q)= \cfrac{1}{\iota \pi}\int\limits_{-\infty}^\infty d\xi \,\,e^{q\sinh \xi -p\xi} \,\,.
\end{eqnarray}
Then, we utilized the following recurrence relations for Hankel functions to simplify the expression:
\begin{eqnarray}\label{eq:Hankel-recurr-rel}
H_{p-1}^{(1)}(q) + H_{p+1}^{(1)}(q) &=& \cfrac{2\,p}{q}\,H_{p}^{(1)}(q)\,\qquad \text{and} \qquad
H_{p-1}^{(1)}(q) - H_{p+1}^{(1)}(q) = 2\, {H_{p}^{(1)}}^\prime(q).
\end{eqnarray}
The remaining non-zero components can similarly be evaluated as:
\begin{eqnarray}\label{eq:Tyy-xy-hyperb}
T_{yy}(\omega^\prime) &=& \int_{-\infty}^{\infty} dt \,e^{\iota \omega^\prime t} \,\mu \,\dot y^2 \,=\, -\iota \,\mu\,\omega^\prime \int_{-\infty}^{\infty} dt \,e^{\iota \omega^\prime t} \, \,y\, \dot y \,= \, -\iota \,\mu \,\omega^\prime \int_{-\infty}^{\infty} d\xi \,e^{\iota \nu(e \sinh \xi -\xi)} \, \,y\, \cfrac{ dy}{d\xi}\nonumber\\
T_{yy}(\nu)
&=&
\mu\,\nu\, \Omega\, a^2 (e^2-1)\, \pi \bigg[ \cfrac{\iota}{\nu e^2}\,{H_{\iota\nu}^{(1)}}(\iota e\nu) + \cfrac{1}{e}\,{H_{\iota\nu}^{(1)}}^\prime(\iota e\nu) \bigg], \nonumber\\
T_{xy}(\omega^\prime) &=& \int_{-\infty}^{\infty} dt \,e^{\iota \omega^\prime t} \,\mu \,\dot x\,\dot y \,=\, -\iota\, \mu\, \omega^\prime \int_{-\infty}^{\infty} dt \,e^{\iota \omega^\prime t} \, \,y\, \dot x \,=\, -\iota\,\mu\,\omega^\prime \int_{-\infty}^{\infty} d\xi \,e^{\iota \nu(e \sinh \xi -\xi)} \, \,x\, \cfrac{ dx}{d\xi} \nonumber\\
T_{xy}(\nu)
&=& -\mu\, \nu\, \Omega\,a^2\,\sqrt{e^2-1}\, \pi\left[ \left( \cfrac{1}{ e^2} - 1 \right)\, {H_{\iota \nu}^{(1)}}(\iota e\nu) + \cfrac{\iota}{\nu e}\,{H_{\iota \nu}^{(1)}}^\prime(\iota e\nu)\right].
\end{eqnarray}
\subsection{Rate of emission of gravitational wave energy}
Using the expressions for $T_{xx}$, $T_{yy}$ and $T_{xy}$ derived in Eqs.~\eqref{eq:Txx-hyperb} and \eqref{eq:Tyy-xy-hyperb} we obtain,
\begin{eqnarray}
T_{ij}(\nu)\,T^*_{ji}(\nu)-\cfrac{1}{3}\,|\,T^{i}{}_{i}(\nu)\,|^2 = \mu^2\,\nu^2\,\Omega^2\,a^4\,\pi^2\,\Big[\,f_1(\nu,e)\,\left(H_{\iota \nu}^{(1)}(\iota e\nu)\right)^2 + f_2(\nu,e)\,\left({H_{\iota \nu}^{(1)}}^\prime(\iota e\nu)\right)^2 \,\Big].
\end{eqnarray}
Here,
\begin{eqnarray}\label{eq:f-nu-e}
f_1(\nu,e) = \cfrac{2}{e^4}\,(e^2-1)^3 + \cfrac{6 - 6e^2 + 2e^4}{3\,\nu^2\,e^4} \qquad \text{and} \qquad f_2(\nu,e) = 2\,\left(\cfrac{e^2-1}{e}\right)\left(\cfrac{1}{\nu^2\,e} + \cfrac{e^2-1}{e}\right).
\end{eqnarray}
After evaluating the integral in Eq.~\eqref{dEdt-2}, we get:
\begin{eqnarray}\label{eq:hyperbolic-power}
P(\nu) & = \cfrac{dE_\text{gw}}{dt} = \cfrac{\kappa^2}{20}\,\,\mu^2\,\Omega^4\,\nu^4\,a^4\,\pi\,\Big[\,\underbrace{f_1(\nu,e)\,\left(H_{\iota \nu}^{(1)}(\iota e\nu)\right)^2 + f_2(\nu,e)\,\left({H_{\iota \nu}^{(1)}}^\prime(\iota e\nu)\right)^2 }_{f(\nu,e)}\,\Big],
\end{eqnarray}
where $f_{1,2}(\nu,e)$ are given in Eq.~\eqref{eq:f-nu-e}. A plot depicting the features of $P(\nu)$ for different values of eccentricity has been shown in Fig.~\ref{fig:power-spectrum}. The total energy loss can be obtained from the power spectrum and is given by:
\begin{eqnarray}
\Delta E = \int_{-\infty}^{\infty} dt \,P(t) = \cfrac{1}{\pi}\int_{0}^{\infty} d\nu\, P(\nu).
\end{eqnarray}
Explicit integration over products of Hankel functions is difficult, especially when the order $p$ of the function $H^{(1)}_p (q)$ also depends on the integration variable. Therefore, we have presented an estimate by doing a numerical integration with various choices of $\nu_\text{max}$ (the upper limit of the integral). The results for different eccentricities have been plotted in Fig.~\ref{fig:energy-radiated}. We notice that, in each case, the curves become parallet to the $x$-axis and approach a constant, for fairly small values of $\nu_\text{max}$.
\begin{figure}[h]
\includegraphics[scale=0.7]{power_spectrum.pdf}
\caption{The power spectrum, in frequency space, of gravitational wave radiation from binaries in hyperbolic orbits with varying eccentricities. }
\label{fig:power-spectrum}
\end{figure}
\begin{figure}[h]
\includegraphics[scale=0.7]{Energy_Loss.pdf}
\caption{Numerical estimate of the total energy loss due to gravitational wave radiation from binaries in hyperbolic orbits. Individual curves correspond to varying eccentricities. The horizontal axis corresponds to the upper limit in the integral $\Delta E = \cfrac{1}{\pi}\int\limits_{0}^{\nu_\text{max}} d\nu \,P(\nu)$. It is evident that even for small enough values of $\nu_\text{max}$, the energy loss attains a constant value. This suggests that with a large $\nu_\text{max}$ the numerical estimate is a good approximation. }
\label{fig:energy-radiated}
\end{figure}
\section{Linear memory from hyperbolic orbits}\label{sec:linear-memory}
\subsection{Power emitted at zero frequency}
Computing the zero frequency limits of the stress tensors is equivalent to obtaining approximations for them in the $\nu$ $\rightarrow$ 0 limit. As $\nu e$ $\rightarrow$ 0, the Hankel function and its first derivative assume the following form:
\begin{eqnarray}\label{eq:Hankel-1-0-lim}
H_{\iota \nu}^{(1)}(\iota e\nu) \simeq \cfrac{2\iota}{\pi} \ln (\nu e), \hspace{1.2cm}
H_{\iota \nu}^{(1)\prime}(\iota e\nu) \simeq \cfrac{2}{\pi \nu e}.
\end{eqnarray}
Substituting the above in the expressions for $T_{xx}$, $T_{yy}$ and $T_{xy}$, in Eqs.~\eqref{eq:Txx-hyperb} and \eqref{eq:Tyy-xy-hyperb} yields,
\begin{eqnarray}\label{eq:T_ij-0-lim-hyperp}
T_{xx} &=& - \cfrac{2\,\mu\, a^{2} \, \Omega}{e^{2}} \bigg[ \ln(\nu e) + (e^{2} - 1) \bigg], \hspace{0.4cm}
\hspace*{1.5cm} T_{yy} = - \cfrac{2\,\mu\, a^{2} \, \Omega}{e^{2}} \left(e^{2} - 1\right) \bigg[ \ln(\nu e) - 1\bigg], \nonumber\\
T_{xy} &=& 2\, \iota \, \mu \, \nu\,\Omega\, a^{2}\, \sqrt{e^{2} - 1} \left[\cfrac{(e^{2} - 1)}{e^{2}} \ln(\nu e) - \cfrac{1}{\nu^{2}\, e^{2}}\right]
\label{TijHype0}
\end{eqnarray}
Using these, we can rewrite,
\begin{eqnarray}
T_{ij}(\nu)\,T^*_{ji}(\nu)-\cfrac{1}{3}\,|\,T^{i}{}_{i}(\nu)\,|^2
= 4\, \mu^{2} \, \nu^2\, a^{4} \left[\widetilde{f}_{1}(\nu, e) (\ln(\nu e))^{2}\, + \, \widetilde{f}_{2}(\nu, e) \ln(\nu e)\, +\, \widetilde{f}_{3}(\nu, e) \right].
\end{eqnarray}
with
\begin{eqnarray}
\widetilde{f}_{1}(\nu, e) &=& \cfrac{2}{e^{4}}(e^{2} - 1)^{3} + \cfrac{6 - 6e^{2} + 2e^{4}}{3\, \nu^{2}\, e^{4}}, \qquad\qquad
\widetilde{f}_{2}(\nu, e) = \cfrac{2\, (e^{2} - 1)}{\nu^{2}\, e^{4}} - \cfrac{6\, (e^{2} - 1)^{2}}{\nu^{2}\, e^{4}}, \nonumber\\
\widetilde{f}_{3}(\nu, e) &=& \cfrac{2\, (e^{2} - 1)}{\nu^{4}\, e^{4}} + \cfrac{2\, (e^{2} - 1)^{2}}{\nu^{2}\, e^{4}}.
\end{eqnarray}
The expression for the power spectrum of the gravitational wave, see Eq.~\eqref{eq:hyperbolic-power}, now becomes,
\begin{eqnarray}
P(\nu) = \cfrac{dE}{dt} = \cfrac{\kappa^{2}}{5}\, \pi\, \mu^{2}\, a^{4}\, \Omega^{4}\, \nu^{4}\, f(\nu,\, e).
\end{eqnarray}
In the limit $\nu \rightarrow 0$,
\begin{eqnarray}
\lim_{\nu \to 0} \nu^{4}\, f(\nu,\, e) = \cfrac{2\, (e^{2} - 1)}{\pi^{2}\, e^{4}},
\end{eqnarray}
which is finite and different from zero, except for $e = 1$ and $e \rightarrow \infty$. Then the power radiated by the GW of zero frequency is given by,
\begin{eqnarray}
P(\nu = 0) = \cfrac{64\, G}{5}\, \mu^{2}\, a^{4}\, \Omega^{4}\, \cfrac{ (e^{2} - 1)}{e^{4}}.
\end{eqnarray}
This result agrees with the expression derived in \cite{Garcia-Bellido:2017knh}.
\subsection{The memory waveform}
The gravitational wave amplitude corresponding to polarisation $\lambda$, measured by a detector located at a distance $r$, is given (in the frequency domain) by
\begin{eqnarray}
h_\lambda(\omega^\prime, r)= \frac{4 G}{r} \epsilon^{ij}_\lambda (\vec n) \, T_{ij}(n, \omega^\prime)
\end{eqnarray}
The waveforms for the $+$ and $\times$ polarisations are
\begin{eqnarray}
h_+ (\omega^\prime, r)&=& \iota\cfrac{4G}{r} \epsilon^{ij}_+(\vec n) T_{ij}(\vec n, \omega^\prime)= \iota\cfrac{4G}{r} \left(\vec{e}_{\theta i} \vec{e}_{\theta j} -\vec{e}_{\phi i} \vec{e}_{\phi j} \right) T_{ij}(\vec n, \omega^\prime)\nonumber\\
&=& \iota \frac{4G}{r} \Big( T_{xx}(\cos^2 \phi -\sin^2\phi \cos^2 \theta) +T_{yy} (\sin^2 \phi -\cos^2\phi \cos^2 \theta) - T_{zz}\sin^2 \theta \nonumber\\
&& \qquad -T_{xy} \sin 2\phi (1+\cos^2\theta) +T_{xz} \sin\phi \sin 2\theta +T_{yz} \cos\phi \sin 2 \theta \Big), \nonumber\\
&&\nonumber\\
h_\times(\omega^\prime,r)&=& \iota\frac{4G}{r} \epsilon^{ij}_\times(\vec n) T_{ij}(\vec n, \omega^\prime)= \iota\frac{4G}{r} \left( \vec{e}_{\theta i} (\vec {e}_\phi)_j + (\vec {e}_\phi)_i (\vec {e}_\theta)_j \right) T_{ij}(\vec n, \omega^\prime)\nonumber\\
&=& \iota \frac{4G}{r} \Big( (T_{xx}-T_{yy}) \sin 2 \phi \cos \theta+ 2 T_{xy} \cos 2\phi\cos \theta) - 2 T_{xz} \cos\phi \sin \theta +2 T_{yz} \sin\phi \sin \theta \Big)\,.
\label{Across-hype}
\end{eqnarray}
We have used the spherical coordinates to describe the polarisation of a GW traversing in the radial $\vec n$ direction. The time domain waveforms can be obtained after evaluating the Fourier transforms of $h_{+} (\omega^\prime, r)$ and $h_{\times} (\omega^\prime, r)$. This can be accomplished numerically and the results have been presented in Fig.~\ref{fig:h-mem-hyperbolic}
\begin{figure}[h]
\centering
\begin{subfigure}
\centering
\includegraphics[scale=0.65]{plusphyp}
\end{subfigure}
\hspace*{0.5cm}
\begin{subfigure}
\centering
\includegraphics[scale=0.65]{crossphyp}
\end{subfigure}
\caption{Plots displaying the characteristics of $h_+(t)$ and $h_\times(t)$ for different eccentricities. Here, $\phi$ = $\theta$ = 0.}
\label{fig:h-mem-hyperbolic}
\end{figure}
\subsection{Comparison with soft-theorems}
According to the soft-graviton theorems, the amplitude of a graviton emission from the scattering of $n$-particles of momenta $p_a$ can be written as the product of a kinematical factor and the amplitude ${\cal A}_{n} (p_a)$ of the n-particle scattering as \cite{Cachazo:2014fwa,Sahoo:2018lxl},
\begin{eqnarray}
{\cal A}_{n+1}(p_a,q)= \frac{\kappa}{2} \epsilon^*_{\lambda \mu \nu} \sum_{a=1}^{n} \Big( \frac{ p_{a\mu} p_{a \nu}}{ p_a\cdot q} -i \frac{ p_{a}^{\mu} q_{ \beta} J_a^{\beta \nu}}{ p_a\cdot q} - \frac{ q_\alpha q_{ \beta} J_a^{\alpha \mu}J_a^{\beta \nu}}{ p_a\cdot q} \Big) {\cal A}_{n} (p_a)
\label{ANLO}.
\end{eqnarray}
Here, $J_a^{\alpha \beta}=x_a^\alpha p_a^\beta-x_a^\beta p_a^\alpha + S_a^{\alpha \beta}$ described the total angular momentum of particle $"a"$. The series of the soft factors are at the same order in the gravitational coupling but in increasing powers of the graviton frequency $q_0=\omega$. The leading term goes as $ \omega^{-1}$ while the sub-leading terms go as $\sim \omega^0$ and $\sim \omega^1$ respectively. The gravitational waveform at distance $r$ of the radiated soft graviton of momentum $q$ is given by
\begin{eqnarray}
h_{\mu \nu}(r,q)= \frac{4G}{r} \Big( \frac{ p_{a\mu} p_{a \nu}}{ p_a\cdot q} -i \frac{ p_{a}^{\mu} q_{ \beta} J_a^{\beta \nu}}{ p_a\cdot q} - \frac{ q_\alpha q_{ \beta} J_a^{\alpha \mu}J_a^{\beta \nu}}{ p_a\cdot q} \Big).
\end{eqnarray}
There are logarithmic corrections to the leading order terms suppressed by $G$ which give a tail contribution to the linear memory signal (which as a function of time goes as $1/t$), \cite{Laddha:2018myi, Sahoo:2018lxl}. The low frequency graviton signal from a generic hard scattering can be written as \cite{Laddha:2018myi},
\begin{eqnarray}
h_{ij}(\omega)= i \omega^{-1} A_{ij} + B_{ij} \log \omega^{-1} +\cdots
\end{eqnarray}
where $A_{ij}\sim {\cal O}({G/r})$ and $B_{ij}\sim {\cal O}({G^2/r})$.
In this section we computed the linear memory for hyperbolic eccentric orbits and it is evident from (\ref{TijHype0}) and (\ref{Across-hype}) that there are logarithmic contributions at ${\cal O}({G/r})$ also.
\section{Non-linear memory }\label{sec:nonlinear-memory}
The non-linear memory is due to the secondary gravitational waves which are emitted by the primary gravitational waves from an oscillating source, such as, a coalescing binary \cite{Thorne:1991, Christodoulou:1991, Will:1991, Favata:2008yd, Favata:2010zu, Tolish:2014bka}.
The stress tensor of gravitational waves is related to the energy radiated as
\begin{eqnarray}
\tau_{ij}^\text{gw}= \cfrac{d E_\text{gw}}{dt \, d\Omega}\, n_i \,n_j ,
\end{eqnarray}
and the non-linear memory waveform is given by \cite{Thorne:1991, Christodoulou:1991,Will:1991},
\begin{eqnarray}
h_{ij}(t, r \,\hat n)= \cfrac{4G}{r}\,\int_{-\infty}^{t-r} \,dt^\prime \,d\Omega^\prime\, \cfrac{dE_\text{gw}}{dt^\prime \,d\Omega^\prime} \,\left[ \cfrac{n^\prime_i \,n^\prime_j}{1-\vec n^\prime \cdot \vec n}\right]^{\rm TT}.
\end{eqnarray}
Here $ \vec n^\prime$ is the unit vector from the source to the solid angle denoted by $d\Omega^\prime$ and $\vec n$ is the unit vector along the line of sight from the source to the detector.
We outline the derivation of these results in this section. The source term of the non-linear gravitational waves is the Landau-Lifshitz stress tensor of the primary gravitational waves given by
\begin{eqnarray}
\tau^\text{gw}_{ij}=\cfrac{1}{32 \pi G} \,\langle \partial_i h^{ab} \partial_j h_{ab} \rangle,
\end{eqnarray}
where the angular brackets encode averaging over time longer than the time period and volumes larger than the wavelengths of the source gravitational waves $h_{ab}$. The source gravitational waves travel outward radially with speed of light and are functions of $(t-r)$, i.e $h_{ab}(t, \vec x)=h_{ab}(t-r, \Omega)$. This implies we can relate their spatial and time derivatives as,
$\partial_i \,h_{ab}(t-r) =-n_i\, \partial_0\, h_{ab}(t-r)$, where $n_i = x_i/r$. Therefore we can write
\begin{eqnarray}\label{source-1}
\tau^\text{gw}_{ij}=\cfrac{1}{32 \pi G} \,\langle \partial_i h^{ab} \partial_j h_{ab} \rangle = n_i n_j \tau^\text{gw}_{00}\,.
\end{eqnarray}
We can model the energy density $\tau^\text{gw}_{00}$ produced from a source and propagating radially outwards on null-rays as
\begin{eqnarray}
\tau^\text{gw}_{00}(t, \vec x) = \cfrac{1}{r^2 }\, \cfrac{dE_\text{gw}(t-r, \Omega)}{dt \,d\Omega}.
\end{eqnarray}
where $dE_\text{gw}/dt$ is the luminosity of the source in gravitational waves and $dE_\text{gw}/d\Omega$ denotes the angular distribution of the source luminosity. Therefore we can write $\tau_{ij}^\text{gw}$ in terms of the energy flux as
\begin{eqnarray}
\tau_{ij}^\text{gw}= n_i n_j \tau_{00}^\text{gw}= n_i n_j \,\cfrac{1}{r^2 } \, \cfrac{dE_\text{gw}(t-r, \Omega)}{dt \,d\Omega}\,.
\end{eqnarray}
The secondary gravitational waves sourced by the gravitational wave stress tensor will obey the in-homogenous wave equation:
\begin{eqnarray}\label{wave-secondary}
\square \,h_{ij}=- 16 \pi G\, \tau^\text{gw}_{ij}.
\end{eqnarray}
whose solution assumes the following form,
\begin{eqnarray}\label{hijtgw-1}
h^\text{mem}_{ij}(t, \vec x)= 4G \int dt^\prime \,d^3 x^\prime\,\, \tau^\text{gw}_{ij} (t^\prime, \vec x^\prime) \, \cfrac{\delta\left(t^\prime -(t-|\vec x -\vec x^\prime|)\right)}{|\vec x -\vec x^\prime|}.
\end{eqnarray}
We can express the source term, Eq.~\eqref{source-1}, in terms of the null coordinate $u=t^\prime -r^\prime$ as follows
\begin{eqnarray}\label{tsource-2}
\tau^\text{gw}_{ij} (t^\prime, \vec x^\prime)&=& \cfrac{n^\prime_i n^\prime_j }{{r^\prime}^2 }\, \cfrac{dE_\text{gw}(t^\prime-r^\prime, \Omega^\prime)}{dt^\prime d\Omega^\prime} \,=\, \int du \,\, \cfrac{n^\prime_i n^\prime_j}{{r^\prime}^2 }\, \, \delta(u-(t^\prime-r^\prime)) \,\cfrac{dE_\text{gw}(u, \Omega)}{dt^\prime d\Omega^\prime}.
\end{eqnarray}
Substituting the above in Eq.~\eqref{hijtgw-1} we obtain
\begin{eqnarray}\label{hijtgw-2}
h^\text{mem}_{ij}(t, \vec x)= 4G \,\int du\, dt^\prime \,dr^\prime\, {r^\prime}^2 \,d\Omega^\prime \, \cfrac{n^\prime_i n^\prime_j}{{r^\prime}^2 }\,\cfrac{\delta\left(t^\prime -(t-|\vec x -\vec x^\prime|)\right) }{|\vec x -\vec x^\prime|} \,\, \delta(u-(t^\prime-r^\prime))\,\cfrac{dE_\text{gw}(u, \Omega)}{dt^\prime d\Omega^\prime}.
\end{eqnarray}
Since the distance to the observer is much larger than the source size, $r \gg r^\prime$, we take the approximations
\begin{eqnarray}
\cfrac{1}{|\vec x-\vec x^\prime|}\simeq \cfrac{1}{r\, (1-\vec n^\prime \cdot \vec n)}\,,\qquad\qquad \delta\left(t^\prime -(t-|\vec x -\vec x^\prime|)\right)\simeq\delta(t^\prime-(t-r)).
\end{eqnarray}
Now we can perform the integral over $r^\prime$ using the second delta function in Eq.~\eqref{hijtgw-2} and then do the $t^\prime$ integration using the remaining delta function to obtain
\begin{eqnarray}\label{hijtgw-3}
h^\text{mem}_{ij}(t, \vec x)= \cfrac{4G}{r} \, \int_{-\infty}^{t-r} \,du \int_{4\pi}\, d\Omega^\prime\,\,\cfrac{dE_\text{gw}(u, \Omega)}{du\, d\Omega^\prime}\,\, \cfrac{n^\prime_i n^\prime_j }{(1-\vec n^\prime \cdot \vec n)}.
\end{eqnarray}
Expression for the corresponding transverse-traceless wavefunction is obtained after multiplication by the projection operator $\Lambda_{ij,kl}(\vec{n})$, defined in Eq.~\eqref{tensor-1},
\begin{eqnarray}\label{hijtgw-4}
\Big[h^\text{mem}_{ij}\Big]^{\text{TT}}(t, \vec x) = \cfrac{4G}{r} \, \int_{-\infty}^{t-r} du \, \int_{4\pi} \, d\Omega^\prime\,\,\cfrac{dE_\text{gw}(u, \Omega)}{du\, d\Omega^\prime}\,\, \cfrac{ \Lambda_{ij,kl}(\vec n)\, n^\prime_k n^\prime_l }{(1-\vec n^\prime \cdot \vec n)},
\end{eqnarray}
\subsection{Non-linear memory from circular orbits}
We will now compute the non-linear memory from binary orbits. The case of elliptical orbits is important from the point of observations as the non-linear memory signal arises at the same Newtonian order as the oscillatory signals. For the case of hyperbolic orbits the non-linear term is suppressed by a factor $(v/c)^5$ compared to the quadrupole signal \cite{Will:1991}.
We will start by computing the non-linear memory from binaries in a circular orbit.
The expression for rate of energy radiated in the direction $d\Omega^\prime$ is given by
\begin{eqnarray}\label{dEdt-m1}
\cfrac{dE_\text{gw}}{dt^\prime\, d\Omega^\prime}=\cfrac{ \kappa^2}{4} \int \, \Big(T_{ij}(\omega^\prime) \,T^*_{kl}(\omega^\prime)\, \Lambda_{ij,kl}(\vec n^\prime) \Big)\,{\omega^\prime}^3\, 2 \pi\, \delta(\omega^\prime-2\omega_0) \, \cfrac{d\omega^\prime}{(2 \pi)^3 \, 2 \omega^\prime}.
\end{eqnarray}
where $\omega_0=\sqrt{G(m_1+m_2)/a^3}$ is the angular frequency of the Kepler orbit.
\begin{figure}[h]
\includegraphics[scale=0.35]{Non-Linear_memory}
\caption{ The axis of rotation of the binary designated by $\vec{L}$ makes an angle $i$ with the $z$-axis and therefore lies in the $y$-$z$ plane. The primary graviton emits a secondary graviton at $\vec r^\prime= r^\prime \hat n^\prime= r^\prime (\sin \theta^\prime \cos\phi^\prime, \sin\theta^\prime \sin \phi^\prime, \cos\theta^\prime)$ and the secondary graviton travels to the earth located at $\vec r= r \hat n=(0,0,1)$.}
\label{fig:Non-Linear-Memory}
\end{figure}
The stress-energy tensor $T_{ij}$, where $i,j = x,y,z$, can be written in matrix form as:
\begin{eqnarray}
\textsf{T}\, (\omega_0) = \cfrac{\mu\, a^2 \omega_0^2}{2}\,\begin{pmatrix}
1 && \iota && 0\\
\iota && -1 && 0\\
0 && 0 && 0
\end{pmatrix}.
\end{eqnarray}
We consider the orientation defined in Fig.~\ref{fig:Non-Linear-Memory}, where the axis of rotation of the binary system lies in the $y$-$z$ plane and is counter-clockwise rotated, by an angle $i$, with respect to the $z$ axis, i.e. $\vec{L} = (0,-\sin i, \cos i)$. Rotation of the axis impacts the form of the stress-tensor matrix \textsf{T} as follows:
\begin{eqnarray}\label{eq:rotation}
\textsf{T} \rightarrow \textsf{T}^\prime = \mathcal{R}\,\textsf{T}\,\mathcal{R}^\text{T}, \hspace{0.2cm} \text{with} \hspace{0.2cm} \mathcal{R} = \begin{pmatrix}
1 && 0 && 0\\
0 && \cos i && -\sin i\\
0 && \sin i && \cos i
\end{pmatrix} \hspace{0.1cm} \Rightarrow \hspace{0.1cm} \textsf{T}^\prime = \cfrac{\mu\, a^2 \omega_0^2}{2} \begin{pmatrix}
1 && \iota \cos i && \iota \sin i\\
\iota \cos i && -\cos^2 i && -\cos i\sin i\\
\iota \sin i && -\cos i\sin i && -\sin^2 i
\end{pmatrix}
\end{eqnarray}
The product $T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\vec n^\prime)$ which can be rewritten entirely as a product of $T_{ij}$'s and $\hat{n}^\prime$, using Eq.~\eqref{tensor-1}, as follows:
\begin{eqnarray}\label{eq:tensor-product-index}
T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime) &=& T_{ij}\,T^*_{ji} - 2\, T_{ij}\,T^*_{jl}\,n_i^\prime\, n_l^\prime + \cfrac{1}{2}\,T_{ij}\, T^*_{kl}\, n_i^\prime\, n_j^\prime\, n_k^\prime\, n_l^\prime + \cfrac{1}{2}\,\left( T_{ii}\,T^*_{kl}\, n_k^\prime\, n_l^\prime + T_{ij}\,T^*_{kk}\, n_i^\prime\, n_j^\prime - T_{ii}\,T^*_{kk} \right).
\end{eqnarray}
For the circular case both the stress-tensor matrices $\mathsf{T}$ and $\mathsf{T}^\prime$ are already traceless. Therefore, the last three terms, within the parentheses, on the right hand side of Eq.~\eqref{eq:tensor-product-index} vanish. The remaining terms can be evaluated as products of matrices as shown below:
\begin{eqnarray}\label{eq:tensor-product-matrix}
T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime) \,&\equiv&\, \text{Tr}\,[\mathsf{T}^\prime\,\mathsf{T}^{\prime\dagger}] - 2\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^\prime\right)\cdot\,\left(\mathsf{T}^{\prime\dagger}\hat{n}^\prime\right) + \cfrac{1}{2}\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^\prime\hat{n}^\prime\right)\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^{\prime\dagger}\hat{n}^\prime\right)
\end{eqnarray}
Substituting for $\mathsf{T}^\prime$, $\hat{n}^\prime$ in Eq.~\eqref{eq:tensor-product-matrix} and evaluating the matrix products and traces yields:
\begin{eqnarray}\label{eq:tensor-product-angles}
T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime) \,&\equiv&\, \cfrac{\mu^2 a^4 \omega_0^4}{4}\, \cfrac{1}{2}\,\Big(1+ 6 \cos^2\theta + \cos^4\theta \Big)\,.
\end{eqnarray}
Here, $\theta$ is the angle between the axis of rotation of the binary $\vec{L} = \left(0,-\sin i, \cos i \right)$ and the direction where the primary graviton emits the second graviton: $\hat n^\prime= \left(\sin \theta^\prime \cos\phi^\prime, \sin\theta^\prime \sin \phi^\prime, \cos\theta^\prime \right)$, i.e.,
\begin{eqnarray}
\cos \theta = -\sin i \sin \theta^\prime \sin \phi^\prime + \cos i \cos \theta^\prime.
\end{eqnarray}
Eq.~\eqref{dEdt-m1} can now be expressed, using the result of Eq.~\eqref{eq:tensor-product-angles} as
\begin{eqnarray}
\cfrac{dE_\text{gw}}{dt^\prime\, d\Omega^\prime}=\cfrac{G}{2\pi}\,\omega_0^6\, \mu^2 a^4\, \Big(1+ 6 \cos^2\theta + \cos^4\theta \Big)\,.
\end{eqnarray}
The Kepler orbital frequency can be written in terms of the semi-major axis $a$, i.e.$\omega_0=(G M/ a^3)^{1/2}$ ($M=m_1+m_2$). This allows us to rewrite the expression for the rate of energy radiated as,
\begin{eqnarray}
\cfrac{dE_\text{gw}}{dt^\prime d\Omega^\prime}=\cfrac{G^4}{2\pi}\cfrac{ \mu^2 M^3}{ a^5} \Big(1+ 6 \cos^2\theta + \cos^4\theta \Big)\,.
\end{eqnarray}
Substituting the above in Eq.~\eqref{hijtgw-3}, we obtain the following expression for the memory signal:
\begin{eqnarray}\label{hijtgw-5}
h_{ij}^\text{mem}(t, \vec x)&=& \cfrac{4G}{r} \int_{-\infty}^{t-r} dt^\prime \int_{4\pi} d\Omega^\prime\,\cfrac{dE_\text{gw}(t^\prime, \Omega)}{dt^\prime d\Omega^\prime} \cfrac{n^\prime_i n^\prime_j }{(1-\hat n^\prime \cdot \hat n)}\nonumber\\
&=& \cfrac{4G}{r} \int_{-\infty}^{t} dt^\prime\, \cfrac{G^4}{2\pi}\,\cfrac{ \mu^2 M^3}{ a^5} \underbrace{\int_{4\pi} d\Omega^\prime\,\Big(1+ 6 \cos^2\theta + \cos^4\theta \Big) \cfrac{n^\prime_i n^\prime_j }{(1-\hat n^\prime \cdot \hat n)}}_{\textsf{A}_{ij}}.
\end{eqnarray}
where we have replaced $t-r$ by $t$ in the limit of the integrand as the waves are detected at a fixed $r$ and the dependence in $u$ then becomes a dependence on time of observation $t$.
To obtain the transverse-traceless part of $h^\text{mem}_{ij}$ we apply the TT projection operator, through multiplication with $\Lambda_{ij,kl}(\vec{n})$, see Eq.~\eqref{hijtgw-4},
\begin{eqnarray}\label{hijtgw-6}
\Big[h_{ij}^\text{mem}\Big]^\text{TT}(t,\vec x)&=& \cfrac{4 G^5}{2 \pi r}\, \mu^2 M^3 \int_{-\infty}^{t} dt^\prime\cfrac{ 1}{ a^5} \, \underbrace{\int_{4\pi} d\Omega^\prime\,\Big(1+ 6 \cos^2\theta + \cos^4\theta \Big) \cfrac{ \Lambda_{ij,kl}(\hat n)\, n^\prime_k n^\prime_l }{(1-\hat n^\prime \cdot \hat n)}}_{\big[\textsf{A}_{ij}\big]^\text{TT}}.
\end{eqnarray}
The tensor product within the angular integral can be further expressed as:
\begin{eqnarray}
\Lambda_{ij,kl}(\hat n)\, n^\prime_k n^\prime_l = \left[ \left(n^\prime_i n^\prime_j -\cfrac{1}{2}\,\delta_{ij} + \cfrac{1}{2}\,n_i n_j \right) - \left( n_i n^\prime_j + n^\prime_i n_j\right)\left( \hat n^\prime \cdot \hat n \right) + \cfrac{1}{2}\,\left(\delta_{ij} + n_i n_j\right)\left( \hat n^\prime \cdot \hat n \right)^2 \right].
\end{eqnarray}
Since, $\hat{n} = (0,0,1)$, $\hat n^\prime \cdot \hat n = \cos \theta^\prime$ and substituting for the components of $\hat{n}$, $\hat{n}^\prime$ and $\delta_{ij}$, the angular integrals can be evaluated, for $i,j = 1,2$ as follows:
\begin{eqnarray}
\Big[\textsf{A}_{11}\Big]^\text{TT} &=& \int_{0}^{\pi} \sin\theta^\prime \,d\theta^\prime\, \int_{0}^{2\pi} d\phi^\prime \,\left(1+ 6 \cos^2\theta + \cos^4\theta \right)\,\times\,\cfrac{1}{2}\left( 1 + \cos \theta^\prime\right)\,\cos 2\phi^\prime = -\cfrac{2\pi}{15}\,\left(17 + \cos^2 i\right) \sin^2 i, \nonumber\\
\Big[\textsf{A}_{12}\Big]^\text{TT} &=& \int_{0}^{\pi} \sin\theta^\prime \,d\theta^\prime\, \int_{0}^{2\pi} d\phi^\prime \,\left(1+ 6 \cos^2\theta + \cos^4\theta \right)\,\times\,\cfrac{1}{2}\left( 1 + \cos \theta^\prime\right)\,\sin 2\phi^\prime = 0.
\end{eqnarray}
The other components are: $ \Big[\textsf{A}_{22}\Big]^\text{TT} = -\Big[\textsf{A}_{11}\Big]^\text{TT}$ and $\Big[\textsf{A}_{12}\Big]^\text{TT} = \Big[\textsf{A}_{21}\Big]^\text{TT} = 0$. The transverse-traceless gravitational wavefunction can be decomposed into $+$ and $\times$ modes:
\begin{eqnarray}\label{eq:hij-mode-decomp}
\Big[\textsf{h}^\text{mem}_{ij}\Big]^\text{TT} = h_I\,\epsilon^I_{ij} = h^\text{mem}_+\,\epsilon^+_{ij} + h^\text{mem}_\times\,\epsilon^\times_{ij}.
\end{eqnarray}
Based on this we can make the identification:
\begin{eqnarray}\label{eq:h-mem-plus-1}
h_{+}^\text{mem}(t,\vec x)&=& \cfrac{4 G^5}{2 \pi r}\,\, \mu^2 M^3 \int_{-\infty}^{t-r} dt^\prime\cfrac{ 1}{ a^5} \, \times \, \left[\cfrac{2\pi}{15}\,\left(17 + \cos^2 i\right) \sin^2 i \right], \qquad h_{\times}^\text{mem}(t,\vec x) = 0.
\end{eqnarray}
The time dependence of the integrand in Eq.~\eqref{hijtgw-5} is due to the change in the radius of the orbit which occurs
due to the energy loss of the orbit by the primary gravitational waves. This change in the radius is given as $da/dt= (dE/dt)\,(da/dE)$, which using $E=-(1/2)\,G \mu M/a $ is
\begin{eqnarray}\label{dadtbinary1}
\cfrac{da}{dt} =-\cfrac{64\, G^3}{5}\, \cfrac{\mu M^2}{a^3}\,.
\end{eqnarray}
Solving this equation gives us the time dependence of the separation distance $a(t)$
\begin{eqnarray}\label{at-1}
a(t)=a_{c}\left[1 + \cfrac{256}{5} \, \cfrac{G^3 \mu M^2}{a_{c}^{4}} (t_c-t) \right]^{1/4}.
\end{eqnarray}
where $t_c$ is the time of coalescence of the black-holes. The growth of memory signal stops at the time of the coalescence $t_{c}$ when the black-hole separation becomes smaller than the innermost stable circular orbit radius $a_c = 6\,GM$. The frequency $f=\omega_0/\pi = (1/\pi) \sqrt{(GM)/a^3}$ increases till the time of coalescence as
\begin{eqnarray}
f(t^\prime)
&\simeq & \left(\cfrac{5}{256}\right)^{3/8}\cfrac{1}{ \pi}\, (G {\mathcal M_c})^{-5/8} \left[\cfrac{5}{256}\, \cfrac{a_{c}^{4}}{G^{3}\mu M^2} + (t_{c} - t^{\prime}) \right]^{-3/8}
\end{eqnarray}
where ${\cal M}_c= \mu^{3/5}M^{2/5}$ is the chirp mass of the binary pair.
Using Eq.~\eqref{at-1} to substitute for $a(t)$ in Eq.~\eqref{eq:h-mem-plus-1}, we obtain the the non-linear memory wavefunction of binaries, whose rotation axis makes an angle $i$ with respect to the earth-source direction, as
\begin{eqnarray}\label{{eq:h-mem-plus-1}}
h^\text{mem}_+(t) & = & \cfrac{1}{192}\cfrac{G}{r}\left(5 G\mu^3 M^{2}\right)^{1/4}\, \frac{1}{\left[5 \left(\frac{3}{2}\right)^4 \frac{GM^2}{\mu} + (t_c - t)\right]^{1/4}} \, {\sin^2 i \left(17+\cos^2 i\right)},\nonumber \\
& = & h_{\text{circ}}\, \cfrac{1}{\left[1 + (t^{\prime}_{c} - t^{\prime}) \right]^{1/4}}\, {\sin^2 i \left(17+\cos^2 i\right)},
\end{eqnarray}
where $h_{\text{circ}} = \frac{2}{3}\, \frac{1}{192}\, \frac{G\mu}{r}$ and $(t^{\prime}_{c} - t^{\prime}) = (t_{c} - t)/T^{\prime}$ with $T^{\prime}$ being $\left(\frac{3}{2}\right)^{4}\frac{5GM^2}{\mu}$.
The memory effect will be strongest in edge on binaries $i=\pi/2$ and will be zero in the face on binaries $i=0$. This is demonstrated in Fig.~\ref{fig:h-mem-vs-t-circular}.
\begin{figure}[!htb]
\includegraphics[scale=0.6]{circularmemory}
\caption{The variation in $h_+^\text{mem}$ against time plotted for different choices of the angle $i$ between the line of sight of observation and the binary axis. }
\label{fig:h-mem-vs-t-circular}
\end{figure}
\section{\bf Elliptical Orbits}\label{sec:elliptical}
For eccentric elliptical orbits, Peters and Mathews \cite{Peters-1,Peters-2} calculated
the average energy and angular momentum emission rates at
Newtonian order. Their calculation has been improved to the 3-PN level \cite{Blanchet:2004ek,Arun:2007sg}, including nonlinear tail-effect (which arises from the scattering of gravitational waves by the near-field potential) at 3-PN \cite{Arun:2007rg}.
In this section we compute the frequency spectrum of the energy radiated following \cite{Mohanty:1994yi,KumarPoddar:2019jxe,KumarPoddar:2019ceq,KumarPoddar:2019ceq,Poddar:2021yjd}. The radiated energy acts as the source term for secondary gravitational waves which carry the non-linear memory signal.
To describe a compact binary system, comprised of stars having masses $m_1$ and $m_2$, in an elliptical orbit, the following quantities are of relevance:
\begin{itemize}
\item Since the orbital motion is around the common centre of mass, it is useful to describe the motion in terms of the reduced mass $\mu = \cfrac{m_1\,m_2}{m_1+m_2}$ and the total mass $M = (m_1 + m_2)$.
\item The elliptical Keplerian orbit can be parametrized as:
\begin{eqnarray}\label{elliptic cordinate}
\quad x(\xi) = a\,(\cos\xi - e),\qquad y(\xi) = b\,\sin\xi, \qquad z(\xi) = 0, \qquad \cfrac{\omega_n^\prime}{n}\,t = \Omega\, t = (\xi - e\sin\xi).
\end{eqnarray}
Here, $a$ and $b \,\,(= a \sqrt{1 - e^2},\,\, e < 1)$ denote the semi-major and semi-minor axes respectively. The variables $e$ and $\xi \in (0, 2\pi)$ refer to the eccentricity of the orbit and the eccentric anomaly respectively.
\item The angular frequency corresponding to the $n^{th}$ harmonic is denoted as $\omega_{n}^\prime = n\,\Omega$, with $n = \{0,1,2,....\}$ being a non-negative integer and the fundamental frequency $\Omega$ can be related to the semi-major axis and the total mass of the system as: $\quad \Omega = \left(\cfrac{G\, (m_1 + m_2)}{a^3}\right)^{1/2}$.
\end{itemize}
The first step in the computation of the rate of energy radiated involves the evaluation of the stress-tensor components in frequency space. Once again starting with the $xx$ component, the calculation proceeds as follows:
\begin{eqnarray}
T_{xx}(\omega_{n}^{\prime}) &=& \cfrac{\mu}{T}\, \int_{0}^{T}\, dt\, \dot{x}^{2}(t) e^{\iota \omega_{n}^{\prime} t}
= \cfrac{-\iota\,\mu\, \omega_{n}^{\prime}}{T}\,\int_{0}^{T}\, dt\, \dot{x}(t)\, x(t)\, e^{\iota \omega_{n}^{\prime} t}.
\end{eqnarray}
In the above, we have used integration by parts. Next, the parametric form of the orbit coordinates, shown in Eq.~\eqref{elliptic cordinate}, allows us to write,
\begin{eqnarray}
\dot{x}\,dt = \cfrac{dx}{d\xi} \,d\xi = - a \sin \xi\, d\xi.
\end{eqnarray}
Using the above transformation and also substituting for $x$ and $\omega_n^\prime\,t$ in terms of functions of $\xi$, we get
\begin{eqnarray}\label{Txx:ellip}
T_{xx}(\omega_{n}^{\prime}) &=& \cfrac{\iota\, \mu a^{2} \omega_{n}^{\prime 2}}{2 \pi n} \int_{0}^{2\pi} d\xi \,\sin\xi \,(\cos\xi - e)\, e^{\iota n (\xi - e \sin\xi)}
= - \cfrac{ \mu a^{2} \omega_{n}^{\prime 2}}{n}\left[\left(\cfrac{1 - e^{2}}{e}\right)J_{n}^{\prime}(ne) - \cfrac{1}{n e^{2}}J_{n}(ne) \right].
\end{eqnarray}
To arrive at the second line of Eq.~\eqref{Txx:ellip}, we replaced the trigonometric functions by the corresponding exponential functions and identified the integral form of Bessel functions of first kind,
\begin{eqnarray}\label{eq:bessel-fn}
J_n(z) = \cfrac{1}{2\pi}\,\int\limits_{0}^{2\pi}\,e^{\iota n (\xi - e \sin\xi)} d\xi.
\end{eqnarray}
The final expression in terms of $J_n(ne)$ and $J^\prime_n(ne)$ is obtained by utilizing the recurrence relations given below:
\begin{eqnarray}\label{eq:bessel-recurr}
J_{n-1}(z) + J_{n+1}(z) = \cfrac{2\,n}{z}\,J_n(z), \hspace{1cm} J_{n-1}(z) - J_{n+1}(z) = 2\,J^\prime_n(z).
\end{eqnarray}
The other non-zero components, i.e $T_{yy}$ and $T_{xy}$ can similarly be obtained as:
\begin{eqnarray}\label{Tyy-xy:ellip}
T_{yy}(\omega_{n}^{\prime}) &=& \cfrac{\mu}{T} \int_{0}^{T} dt\, \dot{y}^{2}(t) e^{\iota \omega_{n}^{\prime} t}
= \cfrac{ \mu\, \omega_{n}^{\prime 2}\, a^{2}(1 - e^{2})}{n}\left[\cfrac{1}{e} J_{n}^{\prime}(ne) - \cfrac{1}{n e^{2}}J_{n}(ne) \right], \nonumber\\
T_{xy}(\omega_{n}^{\prime}) &=& \cfrac{\mu}{T}\, \int_{0}^{T}\, dt\, \dot{x}(t)\dot{y}(t) e^{\iota \omega_{n}^{\prime} t}
= \cfrac{\iota \mu\, \omega_{n}^{\prime 2}\, a^{2}\sqrt{(1 - e^{2})}}{n}\left[-\left(\cfrac{1-e^{2}}{e^{2}}\right)J_{n}(ne) + \cfrac{1}{n e}J_{n}^{\prime}(ne) \right].
\end{eqnarray}
\subsection{Rate of gravitational radiation from binaries in elliptical orbits}
Substituting for $T_{xx}$, $T_{yy}$ and $T_{xy}$ using Eqs.~\eqref{Txx:ellip} and \eqref{Tyy-xy:ellip} gives,
\begin{eqnarray}
T_{ij}(\omega^\prime)\,T^*_{ji}(\omega^\prime)-\cfrac{1}{3}\,|\,T^{i}{}_{i}(\omega^\prime)\,|^2 = g(n, e).
\end{eqnarray}
where we have defined $g(n,e)$ as,
\begin{eqnarray}
g(n,e) &=& J_{n}(ne)^{2}\left[\cfrac{2n^{2}}{e^{4}}(1-e^{2})^{3} + \cfrac{6 - 6e^{2} + 2e^{4}}{3e^{4}} \right] + J_{n}^{\prime}(ne)^{2}\left[\cfrac{2n^{2}}{e^{2}}(1-e^{2})^{2} + \cfrac{2(1 - e^{2})}{e^{2}} \right] \nonumber\\
&& +\, J_{n}(ne)J_{n}^{\prime}(ne) \left[\cfrac{(-8 + 14 e^{2} - 6 e^{4})n}{e^{3}} \right].
\end{eqnarray}
In this case, the energy loss due to gravitational radiation as given in Eq.~\eqref{dEdt-2} becomes,
\begin{eqnarray}
\cfrac{dE_\text{gw}}{dt}(e) &=& \cfrac{\kappa^{2}}{8(2\pi)^{2}} \sum_{n = 0} \cfrac{8\pi}{5}\left[T_{ij}(\omega_{n}^{\prime})T_{ji}^{*}(\omega_{n}^{\prime}) - \cfrac{1}{3}|T^{i}_{}{i}(\omega_{n}^{\prime})|\right] \omega_{n}^{\prime 2}
= \cfrac{32\,G}{20}\,\Omega^{6}\mu^{2}a^{4}\sum_{n = 0}^{\infty} n^{2}g(n,e).
\end{eqnarray}
The series sum over products of Bessel functions and their derivative, weighted by powers of $n$, can be expressed completely as functions of $e$, using the identities derived in the appendix of \cite{Peters-1} and reproduced in Eq.~\eqref{eq:Peter-Matthews-sum}. This enables the identification of the eccentricity dependent part of the energy loss as
\begin{eqnarray}
\widetilde{g}(e) = \sum_{n = 0}^{\infty} n^{2}\,g(n,e) = \cfrac{4}{(1-e^{2})^{7/2}}\left(1 + \cfrac{73}{24}\,e^{2} + \cfrac{37}{96}\,e^{4} \right).
\end{eqnarray}
Therefore, the energy radiated as gravitational waves from the binary can be written as:
\begin{eqnarray}
\cfrac{dE_\text{gw}}{dt} = \cfrac{32\,G}{5}\,\Omega^{6}\left(\cfrac{m_{1}m_{2}}{m_{1}+m_{2}}\right)^{2}a^{4}\cfrac{1}{(1 - e^{2})^{7/2}}\left(1 + \cfrac{73}{24}\,e^{2} + \cfrac{37}{96}\, e^{4} \right).
\end{eqnarray}
\subsection{Non-linear memory from elliptical orbits}
Rate of energy radiated with respect to time and solid angle is described using the following formula:
\begin{eqnarray}
\cfrac{dE_\text{gw}}{dt^\prime \, d\Omega^\prime} &=& \cfrac{\kappa^{2}}{8(2\pi)^{2}} \sum_{n = 0} \bigg[T_{ij}(\omega_n^\prime)\,T^*_{kl}(\omega_n^\prime)\,\Lambda_{ij,kl}(\hat n^\prime)\bigg]\, \omega_{n}^{\prime 2}
\end{eqnarray}
The stress-energy tensor $T_{ij}$, where $i,j = x,y,z$, can be written in matrix form as:
\begin{eqnarray}
\textsf{T}\, (n,e) = \mu\, a^2\, \Omega^2\,\begin{pmatrix}
q_1(n,e) && \iota\,q_2(n,e) && 0\\
\iota\,q_2(n,e) && q_3(n,e) && 0\\
0 && 0 && 0
\end{pmatrix}.
\end{eqnarray}
The non-zero elements of the matrix are functions of $n$, corresponding to the $n$th harmonic, and $e$, the eccentricity of the orbit and these are given as
\begin{eqnarray}
q_1(n,e) &=& -\cfrac{n\,\left(1 - e^{2}\right)}{e}\,J_{n}^{\prime}(ne) + \cfrac{1}{e^{2}}\,J_{n}(ne), \nonumber\\
q_2(n,e) &=& \cfrac{\left(1 - e^{2}\right)^{1/2}}{e}\,J_{n}^{\prime}(ne) - \cfrac{n\,\left(1 - e^{2}\right)^{3/2}}{e^{2}}\,J_{n}(ne), \nonumber\\
q_3(n,e) &=& \cfrac{n\,\left(1 - e^{2}\right)}{e}\,J_{n}^{\prime}(ne) - \cfrac{\left(1 - e^{2}\right)}{e^{2}}\,J_{n}(ne).
\end{eqnarray}
We follow the orientation for the system as shown in Fig.~\ref{fig:Non-Linear-Memory}. The stress-energy matrix for the rotated system is given as $\textsf{T}^\prime = \mathcal{R}\,\textsf{T}\,\mathcal{R}^\text{T}$, with $\mathcal{R}$ being the rotation matrix defined in Eq.~\eqref{eq:rotation}. This allows us to write,
\begin{eqnarray}
\textsf{T}^\prime =\mu\, a^2\, \Omega^2 \begin{pmatrix}
q_1(n,e) && \iota \,q_2(n,e)\, \cos i && \iota \,q_2(n,e)\, \sin i\\
\iota \,q_2(n,e)\, \cos i && q_3(n,e)\, \cos^2 i && q_3(n,e)\, \cos i\sin i\\
\iota \,q_2(n,e)\, \sin i && q_3(n,e)\, \cos i\sin i && q_3(n,e)\, \sin^2 i
\end{pmatrix} .
\end{eqnarray}
Expanding the matrix product $T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime)$ in terms of products of $T_{ij}$ and $n^\prime_i$, yields the same result as Eq.~\eqref{eq:tensor-product-index},
\begin{eqnarray}\label{eq:tensor-product-index-2}
T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime) &=& T_{ij}\,T^*_{ji} - 2\, T_{ij}\,T^*_{jl}\,n_i^\prime\, n_l^\prime + \cfrac{1}{2}\,T_{ij}\, T^*_{kl}\, n_i^\prime\, n_j^\prime\, n_k^\prime\, n_l^\prime + \cfrac{1}{2}\,\left( T_{ii}\,T^*_{kl}\, n_k^\prime\, n_l^\prime + T_{ij}\,T^*_{kk}\, n_i^\prime\, n_j^\prime - T_{ii}\,T^*_{kk} \right).
\end{eqnarray}
Unlike the case of circular orbits, none of the terms on the right hand side of Eq.~\eqref{eq:tensor-product-index-2} vanish. Therefore, the expression when rewritten as a sum over matrix products appears as,
\begin{eqnarray}\label{eq:tensor-product-matrix-2}
T_{ij}\,T^*_{kl}\,\Lambda_{ij,kl}(\hat n^\prime) \,&\equiv&\, \text{Tr}\,[\mathsf{T}^\prime\,\mathsf{T}^{\prime\dagger}] - 2\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^\prime\right)\cdot\,\left(\mathsf{T}^{\prime\dagger}\hat{n}^\prime\right) + \cfrac{1}{2}\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^\prime\hat{n}^\prime\right) \,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^{\prime\dagger}\hat{n}^\prime\right) + \nonumber\\
&& \cfrac{1}{2}\,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^\prime\hat{n}^\prime\right) \,\text{Tr}\,[\mathsf{T}^{\prime\dagger}] + \cfrac{1}{2}\,\text{Tr}\,[\mathsf{T}^\prime] \,\left(\hat{n}^{\prime\text{T}}\,\mathsf{T}^{\prime\dagger}\hat{n}^\prime\right) - \cfrac{1}{2}\,\text{Tr}\,[\mathsf{T}^\prime]\,\text{Tr}\,[\mathsf{T}^{\prime\dagger}].
\end{eqnarray}
Since, $T_{ij}$'s are functions of $n$ and $e$, while $\hat{n}^\prime$ is parametrized in terms of the angles $\theta^\prime$, $\phi^\prime$, the matrix product can be expressed as an overall function of $n$, $e$, $\theta^\prime$, and $\phi^\prime$
\begin{eqnarray}
T_{ij}(n,e) \,T^*_{kl}(n,e)\,\Lambda_{ij,kl}(\theta^\prime, \phi^\prime) &=&\mu^2 a^4\, \Omega^4\, \,\mathcal{I}(n,e,\theta^\prime, \phi^\prime)
\end{eqnarray}
with
\begin{eqnarray}
\mathcal{I}(n,e,\theta^\prime, \phi^\prime) &=&
\cfrac{1}{2}\,q^2_1(n,e)\left[1-p_1^2(\theta^\prime, \phi^\prime)\right]^2 + \cfrac{1}{2}\,q^2_3(n,e)\left[1-p_2^2(\theta^\prime, \phi^\prime)\right]^2 - 2\,q_1(n,e)\,q_3(n,e) \nonumber\\
&&+\, 2\,q^2_2(n,e)\left[1-p_1^2(\theta^\prime, \phi^\prime)\right]\left[1-p_2^2(\theta^\prime, \phi^\prime)\right] + q_1(n,e)\,q_3(n,e)\left[1+p_1^2(\theta^\prime, \phi^\prime)\right]\left[1+p_2^2(\theta^\prime, \phi^\prime)\right].
\end{eqnarray}
Here,
\begin{eqnarray}
p_1(\theta^\prime, \phi^\prime) = \sin\theta^\prime\,\cos\phi^\prime \quad \text{and} \quad p_2(\theta^\prime, \phi^\prime) = \cos i \sin \theta^\prime \sin \phi^\prime + \sin i \cos \theta^\prime
\end{eqnarray}
Computation of the transverse-traceless wavefunction involves (i) an angular integral over $\theta^\prime$, $\phi^\prime$ and (ii) a sum over the orders ($n$) of the Bessel functions. This will ultimately yield an eccentricity dependent function:
\begin{eqnarray}
\Big[\textsf{A}_{ij}\Big]^\text{TT}(e) = \sum_{n = 0}^{\infty}\, \, \int_{4\pi} d\,\Omega^\prime\,\mathcal{I}(n,e,\theta^\prime, \phi^\prime)\, \cfrac{ \Lambda_{ij,kl}(\hat n)\, n^\prime_k n^\prime_l }{(1-\hat n^\prime \cdot \hat n)}.
\end{eqnarray}
The tensor product within the angular integral can be further expanded as:
\begin{eqnarray}
\Lambda_{ij,kl}(\hat n)\, n^\prime_k n^\prime_l = \left[ \left(n^\prime_i n^\prime_j -\cfrac{1}{2}\,\delta_{ij} + \cfrac{1}{2}\,n_i n_j \right) - \left( n_i n^\prime_j + n^\prime_i n_j\right)\left( \hat n^\prime \cdot \hat n \right) + \cfrac{1}{2}\,\left(\delta_{ij} + n_i n_j\right)\left( \hat n^\prime \cdot \hat n \right)^2 \right].
\end{eqnarray}
Since, $\hat{n} = (0,0,1)$, $\hat n^\prime \cdot \hat n = \cos \theta^\prime$ and substituting for the components of $\hat{n}$, $\hat{n}^\prime$ and $\delta_{ij}$, the angular integrals can be evaluated, for $i,j = 1,2$ as follows:
\begin{eqnarray}
\Big[\textsf{A}_{11}\Big]^\text{TT}(e) &=& \sum_{n = 0}^{\infty}\, \, \int_{0}^{\pi} \sin\theta^\prime \,d\theta^\prime\, \int_{0}^{2\pi} d\phi^\prime \, \mathcal{I}(n,e,\theta^\prime, \phi^\prime) \,\times\,\cfrac{1}{2}\left( 1 + \cos \theta^\prime\right)\,\cos 2\phi^\prime \nonumber\\
&=& \cfrac{2\pi}{15}\,\sum_{n = 0}^{\infty}\, \bigg[ \Big( 6\,q_1(n,e)\,q_3(n,e) -3\,q^2_1(n,e) - 8\,q^2_2(n,e) \Big) \nonumber\\
&& \qquad\qquad + \Big( 8\,q^2_2(n,e) + 2\,q^2_3(n,e) -6\,q_1(n,e)\,q_3(n,e) \Big) \cos^2 i + q^2_3(n,e) \cos^4 i\bigg] \nonumber \\
& = & \cfrac{2\pi}{15} \left(C_{0}(e) + C_{2}(e)\cos^{2} i + C_{4}(e)\cos^{4} i\right)
\end{eqnarray}
The sum over $n$ can be evaluated, using the identities given in Eq.~\eqref{eq:Peter-Matthews-sum}, to obtain the coefficients $C_{i}(e)$, $i = 0,1,2$:
\begin{eqnarray}
C_{0}(e) & = & \sum_{n = 0}^{\infty}\,\bigg[ \cfrac{(6e^{2} - 9)}{e^{4}}\, n^{2} \left[J_{n}(ne)\right]^2 - \cfrac{8(1-e^{2})^{3}}{e^{4}}\, n^{4} \left[J_{n}(ne)\right]^2 - \cfrac{8(1-e^{2})}{e^{2}}\, n^{2} \left[J^\prime_{n}(ne)\right]^2 - \cfrac{9(1-e^{2})^{2}}{e^{2}}\, n^{4} \left[J^\prime_{n}(ne)\right]^2 \nonumber \\
& & \qquad\qquad + \cfrac{(1-e^{2})}{e^{3}}( 34 - 22e^{2})\, n^{3} \left[J_{n}(ne)\,J^\prime_{n}(ne)\right] \bigg] \nonumber \\
& = & - \cfrac{1}{4 e^{2}(1 - e^{2})^{7/2}} \,\bigg[68\, e^{2} + \cfrac{1565}{8}\, e^{4} + \cfrac{1071}{64}\, e^{6} \bigg],
\nonumber\\
&&\nonumber\\
C_{2}(e) &=& \sum_{n = 0}^{\infty}\,\bigg[ \cfrac{2(4-5e^{2} + e^{4})}{e^{4}}\, n^{2} \left[J_{n}(ne)\right]^2 + \cfrac{8(1-e^{2})^{3}}{e^{4}}\, n^{4} \left[J_{n}(ne)\right]^2 + \cfrac{8(1-e^{2})}{e^{2}}\, n^{2} \left[J^\prime_{n}(ne)\right]^2 + \cfrac{8(1-e^{2})^{2}}{e^{2}}\, n^{4} \left[J^\prime_{n}(ne)\right]^2 \nonumber \\
& & \qquad\qquad + \cfrac{2(1-e^{2})}{e^{3}}(-16 + 13 e^{2})\, n^{3} \left[J_{n}(ne)\,J^\prime_{n}(ne)\right] \bigg] \nonumber\\
&=& \cfrac{1}{4 e^{2}(1 - e^{2})^{7/2}} \bigg[64\, e^{2} + \cfrac{399}{2}\, e^{4} + \cfrac{101}{4}\, e^{6} \bigg], \nonumber\\
&& \nonumber\\
C_{4}(e) &=& \sum_{n = 0}^{\infty}\, \cfrac{(1-e^{2})^{2}}{e^{4}} \bigg[ n^{2} \left[J_{n}(ne)\right]^2 + e^{2} n^{4} \left[J^\prime_{n}(ne)\right]^2 - 2 e\, n^{3} \left[J_{n}(ne)\,J^\prime_{n}(ne)\right] \bigg] \nonumber\\
&=& \cfrac{1}{4e^{2}(1 - e^{2})^{9/2}} \bigg[ -1 - \cfrac{15}{4}\, e^{2} + \cfrac{83}{4}\, e^{4} + \cfrac{21}{8}\, e^{6} - \cfrac{37}{64}\, e^{8} \bigg].
\end{eqnarray}
The cross-component $\big[\textsf{A}_{12}\big]^\text{TT}(e)$ can similarly be evaluated as:
\begin{eqnarray}
\Big[\textsf{A}_{12}\Big]^\text{TT}(e) &=& \sum_{n = 0}^{\infty}\, \, \int_{0}^{\pi} \sin\theta^\prime \,d\theta^\prime\, \int_{0}^{2\pi} d\phi^\prime \, \mathcal{I}(n,e,\theta^\prime, \phi^\prime) \,\times\,\cfrac{1}{2}\left( 1 + \cos \theta^\prime\right)\,\sin 2\phi^\prime = 0.
\end{eqnarray}
The other components can be obtained using the above as: $ \big[\textsf{A}_{22}\big]^\text{TT} = -\big[\textsf{A}_{11}\big]^\text{TT}$ and $\big[\textsf{A}_{12}\big]^\text{TT} = \big[\textsf{A}_{21}\big]^\text{TT} = 0$. The decomposition of the transverse-traceless wavefunction into $+$ and $\times$ polarization modes, see Eq.~\eqref{eq:hij-mode-decomp}, allows us to identify:
\begin{eqnarray}\label{eq:wavefunc-ellip}
h_{+}^\text{mem}(t,\vec x)&=& \cfrac{4G}{r} \int_{-\infty}^{t-r} dt^\prime\cfrac{ 1}{\pi} \cfrac{G^4 \mu^2 M^3}{a(t^\prime)^5} \, \times \, \cfrac{2\pi}{15}\, \Big(C_{0}(e) + C_{2}(e) \cos^2 i + C_{4}(e) \cos^4 i\Big), \qquad h_{\times}^\text{mem}(t,\vec x) = 0.
\end{eqnarray}
Within the integrand, the time-dependence is encoded in both the semi-major axis $a$ and the eccentricity $e$ of the orbit. This is on account of the fact that energy loss due to primary gravitational waves alters the features of the orbit.
The explicit time-dependence for $a(t)$ and $e(t)$ can be obtained by solving the following system of non-linear differential equations \cite{Peters-2}:
\begin{eqnarray}\label{dadt-dedt}
&& \cfrac{da}{dt} = - \cfrac{64}{5}\,\cfrac{G^{3}\mu M^{2}}{a^{3}\,(1 - e^2)^{7/2}}\,\left( 1 + \cfrac{73}{24}\,e^2 + \cfrac{37}{96}\,e^4 \right), \\
&& \cfrac{de}{dt} = - \cfrac{304}{15}\,\cfrac{G^{3}\mu M^{2}}{a^{4}\,(1 - e^2)^{5/2}}\,\left( 1 + \cfrac{121}{304}\,e^2 \right).
\end{eqnarray}
The above can be used to eliminate $t$ and obtain a differential equation involving $a$ and $e$, i.e.,
\begin{eqnarray}\label{eq:a(e)}
\cfrac{da}{de} = \cfrac{12}{19}\,\cfrac{a}{e\,(1-e^2)}\,\cfrac{1 + (73/24)\,e^2 + (37/96)\,e^4}{1 + (121/304)\,e^2},
\end{eqnarray}
whose solution expresses $a$ in terms of $e$,
\begin{eqnarray}
a(e) = c_{0} \,\cfrac{e^{12/19}}{(1-e^2)}\left(1 + \cfrac{121}{304}\,e^2\right)^{870/2299},
\end{eqnarray}
where $c_{0} = a_{0}\, e_{0}^{-12/19}(1-e_{0}^{2})(1 + (121/304)\,e_{0}^{2})^{-870/2299}$ and $a_{0} = a(e_{0})$ defines the initial condition necessary for solving the differential equation.
The time evolution can be given as a function of eccentricity as
\begin{eqnarray}\label{eq:t-tc(e)}
t - t_{c} = -\cfrac{15}{304}\,\cfrac{c_{0}^{4}}{G^{3}\mu M^{2}}\, \cfrac{19}{48}\,e^{48/19}\,F_{1}\left(\cfrac{24}{19}\,, -\cfrac{1181}{2299}\,, \cfrac{3}{2}\,, \cfrac{43}{19}\,; -\cfrac{121}{304}\,e^2, e^2 \right),
\end{eqnarray}
where $t_c$ corresponds to the instant of coalescence of the black-holes, i.e., when $e$ becomes close to zero. $F_{1}$ is the hypergeometric Appell function with the following integral representation:
\begin{eqnarray}
F_{1}(a, b_{1}, b_{2}, c; x, y) = \cfrac{\Gamma(c)}{\Gamma(a)\Gamma(c-a)}\int_{0}^{1} t^{a-1}(1-t)^{c-a-1}(1-xt)^{-b_{1}}(1-yt)^{-b_{2}}dt,\,\,\,\,\,\,\,\, \mathfrak{R}c > \mathfrak{R}a > 0.
\end{eqnarray}
Eq.~\eqref{eq:t-tc(e)} can be expressed entirely in terms of dimensionless quantities by noting that $\overline{T} = c_0^4/ (G^{3}\mu M^2)$ has the dimensions of time. Therefore,
\begin{eqnarray}\label{eq:t-tc(e)-dimless}
\cfrac{t - t_{c}}{\overline{T}} = \overline{t} - \overline{t}_{c} = -\cfrac{15}{304}\, \cfrac{19}{48}\,e^{48/19}\,F_{1}\left(\cfrac{24}{19}\,, -\cfrac{1181}{2299}\,, \cfrac{3}{2}\,, \cfrac{43}{19}\,; -\cfrac{121}{304}\,e^2, e^2 \right),
\end{eqnarray}
The wavefunction in Eq. (\ref{eq:wavefunc-ellip}) can be rewritten, after a change of integration variable, from $t^\prime$ to $e^\prime$ as,
\begin{eqnarray}\label{eq:h_+(e)}
h_{+}^\text{mem}(e,\vec x) &=& \cfrac{4G}{r}\int_{e}^0 \cfrac{de^\prime}{\dot{e}(e^\prime)}\cfrac{1}{\pi}\cfrac{G^4 \mu^2 M^3}{a(e^\prime)^5}\,\cfrac{2\pi}{15}\left[C_{0}(e^\prime) + C_{2}(e^\prime)\cos^2 i + C_{4}(e^\prime)\cos^4 i \right],\nonumber \\
& = & \cfrac{4G}{r}\cfrac{15}{304}\cfrac{G \mu M}{c_{0}}\, \cfrac{2}{15} \left[C_{h}^{(0)}(e) + C_{h}^{(2)}(e)\cos^2 i + C_{h}^{(4)}(e)\cos^4 i \right],
\end{eqnarray}
or in terms of dimensionless quantities, we can write
\begin{eqnarray}\label{eq:h_+(e)-dimless}
\cfrac{h_{+}^\text{mem}(e,\vec x)}{h_0} & = & \overline{h}_{+}^\text{mem}(e,\vec x) = \cfrac{1}{152} \left[C_{h}^{(0)}(e) + C_{h}^{(2)}(e)\cos^2 i + C_{h}^{(4)}(e)\cos^4 i \right].
\end{eqnarray}
Here, $h_0 = (4 \,G^2 \mu M)/(r\, c_0)$. The eccentricity dependent part is constituted of
\begin{eqnarray}
C_{h}^{(0)}(e) &=& 17\,\cfrac{19 \,e^{-12/19}}{12}\,F_{1}\left(-\cfrac{6}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{13}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right) - \cfrac{1565}{32}\,\cfrac{19\, e^{26/19}}{26}\,F_{1}\left(\cfrac{13}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{32}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right) \nonumber \\&&- \cfrac{1071}{256}\,\cfrac{19\, e^{64/19}}{64}\,F_{1}\left(\cfrac{32}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{51}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right),\nonumber \\
C_{h}^{(2)}(e) &=& -16\,\cfrac{19 \,e^{-12/19}}{12}\,F_{1}\left(-\cfrac{6}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{13}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right) + \cfrac{399}{8}\,\cfrac{19\, e^{26/19}}{26}\,F_{1}\left(\cfrac{13}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{32}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right) \nonumber \\&& + \cfrac{101}{16}\,\cfrac{19\, e^{64/19}}{64}\,F_{1}\left(\cfrac{32}{19}\,, \cfrac{3169}{2299}\,, 0\,, \cfrac{51}{19}\,; -\cfrac{121}{304}\,e^2\,,0 \right),\nonumber \\
C_{h}^{(4)}(e) &=& \cfrac{1}{4}\,\cfrac{19 \,e^{-50/19}}{50}\,F_{1}\left(-\cfrac{25}{19}\,, \cfrac{3169}{2299}\,, 1\,, -\cfrac{6}{19}\,; -\cfrac{121}{304}\,e^2\,,e^2 \right) + \cfrac{15}{16}\,\cfrac{19\, e^{-12/19}}{12}\,F_{1}\left(-\cfrac{6}{19}\,, \cfrac{3169}{2299}\,, 1\,, \cfrac{13}{19}\,; -\cfrac{121}{304}\,e^2\,, e^2 \right) \nonumber \\ && + \cfrac{83}{16}\,\cfrac{19\, e^{26/19}}{26}\,F_{1}\left(\cfrac{13}{19}\,, \cfrac{3169}{2299}\,, 1\,, \cfrac{32}{19}\,; -\cfrac{121}{304}\,e^2\,, e^2 \right) + \cfrac{21}{32}\,\cfrac{19\, e^{64/19}}{64}\,F_{1}\left(\cfrac{32}{19}\,, \cfrac{3169}{2299}\,, 1\,, \cfrac{51}{19}\,; -\cfrac{121}{304}\,e^2\,, e^2 \right)\nonumber \\ && -\cfrac{37}{256}\,\cfrac{19\, e^{102/19}}{102}\,F_{1}\left(\cfrac{51}{19}\,, \cfrac{3169}{2299}\,, 1\,, \cfrac{70}{19}\,; -\cfrac{121}{304}\,e^2\,, e^2 \right),
\end{eqnarray}
The variation in the memory waveform with time can be obtained by exploiting the relation between the waveform and eccentricity of the orbit, as described in Eq.~\eqref{eq:h_+(e)-dimless} and the implicit relation between time and eccentricity highlighted in Eq.~\eqref{eq:t-tc(e)-dimless}. A plot depicting such variation is shown in Fig.~\ref{fig:h+-mem-vs-t-tc}.
\begin{figure}[h]
\centering
\begin{subfigure}
\centering
\includegraphics[scale=0.42]{GW_ellip1}
\end{subfigure}
\begin{subfigure}
\centering
{\includegraphics[scale=0.42]{GW_ellip2}}
\end{subfigure}
\caption{ The variation in $\overline{h}_+^\text{mem}$ with respect to the time interval $\overline{t}-\overline{t}_c$ plotted for different choices of the angle between between the line of sight of observation and the binary axis. The first figure displays the $\overline{h}_+^\text{mem}$ values on a logarithmic scale, whereas the second figure presents the same information but with the regular scaling of the axes.}
\label{fig:h+-mem-vs-t-tc}
\end{figure}
An alternative description can be obtained in terms of the frequency $\nu = (2\pi)^{-1}\,m^{1/2}\,a^{-3/2}$, by noting that the relation between $\nu$ and $e$ can be obtained by solving the following differential equation:
\begin{eqnarray}\label{eq:dnu-de}
\cfrac{d\nu}{de} = -\cfrac{18}{19}\,\cfrac{\nu}{e\,(1-e^2)}\,\cfrac{1 + (73/24)\,e^2 + (37/96)\,e^4}{1 + (121/304)\,e^2}.
\end{eqnarray}
The solution can be evaluated as,
\begin{eqnarray}\label{eq:nu(e)}
\nu(e) = \widetilde{c}_{0} \,e^{-18/19}\,\left(1-e^2\right)^{3/2}\left(1 + \cfrac{121}{304}\,e^2\right)^{-1305/2299},
\end{eqnarray}
where $\widetilde{c}_{0} = \nu_{0}\, e_{0}^{-18/19}(1-e_{0}^{2})^{-3/2}(1 + (121/304)\,e_{0}^{2})^{1305/2299}$ with $\nu_{0} = \nu(e_{0})$ defining the initial condition. In terms of dimensionless quantities, the same equation can be rewritten as,
\begin{eqnarray}\label{eq:nu(e)-dimless}
\cfrac{\nu(e)}{\widetilde{c}_{0}} = \overline{\nu}(e) = e^{-18/19}\,\left(1-e^2\right)^{3/2}\left(1 + \cfrac{121}{304}\,e^2\right)^{-1305/2299},
\end{eqnarray}
The variation in the memory signal with respect to the frequency can be obtained based on their mutual dependence on eccentricity, as highlighted in Eqs.~\eqref{eq:h_+(e)-dimless} and \eqref{eq:nu(e)-dimless}. A plot depicting such a relation is shown in Fig.~\ref{fig:h+-mem-vs-frequency}.
\begin{figure}[!htb]
\includegraphics[scale=0.55]{GW_ellip3}
\caption{The variation in $\overline{h}_+^\text{mem}$ with respect to the dimensionless frequency parameter $\bar{\nu}$, defined in Eq.~\eqref{eq:nu(e)-dimless}, plotted for different choices of the angle between between the line of sight of observation and the binary axis.}
\label{fig:h+-mem-vs-frequency}
\end{figure}
\section{\bf Conclusions}\label{sec:conclusion}
In this paper we have computed the linear memory signal for eccentric
hyperbolic encounters in both frequency and time domain. We performed a field theoretic calculation of the amplitude of graviton emission from a classical stress tensor and related this amplitude to the gravitational wave signal as a function of frequency. We took the zero-frequency limit of the waveform to identify the memory signals which we computed in the frequency space and taking the Fourier transform also in the time domain. We performed these calculations for hyperbolic orbits with large eccentricity while retaining terms of all orders in eccentricity. We find that in the eccentric hyperbolic orbits the low-frequency memory component has terms which vary with the frequency $\omega=\nu \Omega$ of the gravitational waves as $\log (\omega e/\Omega)$. This $\log$ dependence is due to the eccentric hyperbolic orbit and is different from the tail terms which occur for any unbound scattering at ${\cal O}(G^2)$ \cite{Laddha:2018myi,Sahoo:2018lxl,Laddha:2018vbn}. Hyperbolic encounters of supermassive black-holes of mass $10^7 M_{\odot}$ with closest approach of 1pc will produce gravitational waves with peak frequency $\sim 0.2$ mHz and encounters between blackholes of mass $10^3 M_\odot$ will produce gravitational waves at peak frequency $0.04$ mHz . The time window for the observations is of the order of 1 day which makes such observations possible for third generation gravitational detectors.
The non-linear memory which occurs from GW radiated by GW is equally important as it would prove the non-linear nature of graviton-graviton interactions. We have computed the non-linear memory for eccentric elliptical orbits. Highly eccentric orbits are possible if the initial system has three bodies of which one is ejected \cite{Heggie} or by capture from an unbound orbit \cite{Quinlan, OLeary:2008myb}. We compute the frequency spectrum energy radiated by the binaries using the field theory technique which gives the energy spectrum directly in the frequency space. This is used as the source term for the secondary gravitational waves, which results in the memory waveform in the frequency space. We use the instantaneous eccentricity as the parameter for tracking the change of frequency and semi-major axis in time due to radiation reaction. We thus obtain the non-linear memory as a function of the instantaneous eccentricity. We then exploited the change in eccentricity with respect to both time and frequency to obtain the non-linear waveform as a function of both time and frequency. The calculation of the non-linear memory has been done up to all orders in eccentricity. These signal templates in frequency and time domain may be useful for extracting the memory signal from data by the future detectors like Advanced-LIGO, Einstein Telescope and LISA.
\section{\bf Appendix}\label{sec:appendix}
For evaluating the sum over $n$ for elliptical binaries, we have used following formulae \cite{Peters-1}:
\begin{eqnarray}\label{eq:Peter-Matthews-sum}
&& \sum_{n=0}^{\infty} \, n^{2}J_{n}^{2}(ne) = \cfrac{e^{2}}{4(1-e^{2})^{7/2}}\left(1 + \cfrac{e^{2}}{4} \right), \nonumber \\
&& \sum_{n=0}^{\infty} \, n^{2}[J_{n}^{\prime}(ne)]^{2} = \cfrac{e^{2}}{4(1-e^{2})^{5/2}}\left(1 + \cfrac{3e^{2}}{4} \right), \nonumber \\
&& \sum_{n=0}^{\infty} \, n^{3}J_{n}(ne) J_{n}^{\prime}(ne) = \cfrac{e}{4(1-e^{2})^{9/2}}\left(1 + 3e^{2} + \cfrac{3}{8}e^{4} \right), \nonumber \\
&& \sum_{n=0}^{\infty} \, n^{4}J_{n}^{2}(ne) = \cfrac{e^{2}}{4(1-e^{2})^{13/2}}\left(1 + \cfrac{37}{4}e^{2} + \cfrac{59}{8}e^{4} + \cfrac{27}{64}e^{6} \right), \nonumber \\
&& \sum_{n=0}^{\infty} \, n^{4}[J_{n}^{\prime}(ne)]^{2} = \cfrac{1}{4(1-e^{2})^{11/2}}\left(1 + \cfrac{39}{4}e^{2} + \cfrac{79}{8}e^{4} + \cfrac{45}{64}e^{6} \right).
\end{eqnarray}
\begin{acknowledgements}
S.P. is supported by the MHRD, Government of India, under the Prime Minister's Research Fellows (PMRF) Scheme, 2020. A.H. would like to thank MHRD, Government of India for research fellowship.
\end{acknowledgements}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 6,270
|
Margaretaskolorna var svenska hushållsskolor som byggde på ett nytt koncept. De bestod av servering, utbildningsverksamhet, butik med färdiglagad mat och ibland även festvåning. Den första Margaretaskolan öppnade 1905 i Norrköping och den sista Margaretaskolan stängdes 1977. Den svenska entreprenören Hanna Lindmark stod bakom Margaretaskolorna och vid hennes död 1941 testamenterades verksamheten till fyra missionssamfund, Svenska kyrkans mission, Svenska missionsförbundet, "Svenska missionen i Kina" och Sällskapet Svensk Baptistmission. Verksamheten drevs vidare i bolagsform av de fyra samfunden, men avvecklades under 1970-talet. År 1976-1992 var Svenska Missionsförbunde ensam ägare av bolaget samt ansvarig för hushållsskolans pensionsstiftelse.
Första Margaretaskolan i Norrköping
Hanna Lindmark öppnade sin första matservering i Östersund. I samband med öppnandet av hushållsskolan i Norrköping skapades benämningen Margaretaskolan. En bidragande orsak till etableringen var den kommande konst- och industriutställningen i Norrköping 1906 vilket de räknade med skulle skapa efterfrågan på serveringsställen.
Koncept
Det fanns behov av städade serveringar (dit även kvinnor och barn kunde gå) och det fanns behov för unga kvinnor att lära sig matlagning. Hanna Lindmark kombinerade dessa i Margaretaskolorna där unga kvinnor fick betala för att gå i hushållsskola och fick öva genom att bedriva verksamheten: servering, butik med färdiglagad mat och festvåning. Margaretaskolorna vilade på kristen grund. Etablissemangen var alkoholfria och var mycket ansedda.
Etableringar
Sammanlagt öppnade 12 Margaretaskolor: Norrköping, Stockholm (där den första av fyra etableringar startade på Bältgatan 5 lagom till olympiska sommarspelen 1912, andra adresser var Regeringsgatan 47, Citypalatset på Norrmalmstorg 1-3 och Fenixpalatset på Adolf Fredriks kyrkogata 10), Göteborg (Dicksonska palatset), Helsingborg, Linköping (servering vid stora torget, butik på Bokhållaregatan), Örebro, Borås, Västerås, Lund, Malmö och Jönköping.
Nedgång och fall
Den sista Margaretaskolan stängdes 1977 efter en lång tids nedgång. Efter Hanna Lindmarks död övergick ägandet av Margaretaskolorna till bland annat Missionsförbundet. Oscar von Malmborg tog kontroll över bolaget och hans svärson Gunnar Törnqvist blev VD. Den senare förde ett utsvävande liv, vilket bekostades med förskingrade medel från Margaretaskolorna.
Källor
Noter
Litteratur
Ewonne Winblad: Frälst, förmögen, förskingrad (2008)
Hushållsskolor
Skolformer
Utbildning i Sverige
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 8,696
|
\section{Introduction}
\label{sec:intro}
One of the most fundamental goals of galaxy evolution studies is the robust measurement of the cosmic history of star formation \citep{Madau_2014}, the most direct tracer of which is the rest-frame UV light, dominated by the emission from massive stars \citep{Kennicutt_2012}. The UV luminosity, with the aid of the assumed stellar initial mass function (IMF), is therefore the most widely used observable in determining the corresponding star formation, particularly at high redshifts (e.g.\ \citealt{Bouwens_2011, Dunlop_2012, Ellis_2013, Oesch_2014, Oesch_2015, Bouwens_2015, McLeod_2015, McLeod_2016}). However, the use of the UV luminosity as a tracer of the star-formation rate (SFR) is hampered by the absorption of the UV light by the dust present in the interstellar medium (ISM). The energy absorbed by dust is re-emitted in the infrared (IR) and, therefore, one ideally requires the IR observations to complement the UV data, in order to paint complete picture of star formation.
At $z>2$, most studies of the evolution of the star-formation rate density (SFRD) are based on samples of Lyman-break galaxies (LBGs), due in part because of the efficiency of their selection technique in deep broad-band imaging surveys. As a result, LBGs have been extensively studied and well characterised over the past two decades. With stellar masses of $\sim 10^{9-11}\,{\rm M_\odot}$ and star-formation rates (SFRs) of $\sim$\,$10$--$100$\,${\rm M_\odot\,yr^{-1}}$ (e.g.\ \citealt{Madau_1996, Reddy_2006, Stark_2009, Oteo_2013}), LBGs have been suggested to be responsible for forming a substantial fraction of the stellar mass in massive local galaxies ($L\geq {\rm L_\ast}$; e.g.\ \citealt{Somerville_2001, Baugh_2005}), with those with the highest SFRs ($>100\,{\rm M_\odot\, yr^{-1}}$) being potentially the progenitors of present-day ellipticals (e.g.\ \citealt{Verma_2007, Reddy_2009, Stark_2009}). However, the influence of dust extinction is expected to have a significant impact on the derived estimates of the star-formation rates of individual sources, as well as the completeness of these UV-selected studies. Hence, it is important to try to constrain the far-infrared (FIR) luminosities of LBGs to derive more robust estimates of their SFRs. Unfortunately, due to the relatively modest sensitivity of the single-dish far-infrared and sub-/millimetre facilities and the rather small fields of view of sub/-millimetre interferometers, the vast majority of LBGs lack strong constraints on their rest-frame far-infrared or sub-millimetre emission and hence alternative methods have to be used in order to account for the dust-absorbed UV light (e.g. \citealt{Chapman_2000, Chapman_2002b, Webb_2003, Coppin_2015}).
The most common approach to estimating the far-infrared luminosity of UV-selected galaxies is to use the observed correlation between the so-called infrared excess, ${\rm IRX}\equiv L_{\rm IR}/L_{\rm UV}$, and the measured UV slope, $\beta$, (e.g.\ \citealt{Meurer_1999, Overzier_2011, Takeuchi_2012}). Dust obscuration becomes progressively less important towards near-IR wavelengths (dust extinction is about ten times lower at 2.2\,${\rm \mu m}$ than at 0.55\,${\rm \mu m}$; \citealt{Calzetti_1997}), so that the slope of the rest-frame UV stellar emission spectrum, $\beta$, can be used as an indication of the amount of dust attenuation (with $F_\lambda\propto \lambda^\beta$). Although it should be stressed that this estimate only applies to the {\it detected} UV continuum emission (if emission is too highly obscured then it will be completely missed) and is weighted towards the least extincted parts of the source.
It has been found that IRX correlates tightly with the observed UV slope for local UV-selected starburst galaxies \citep{Calzetti_1997, Meurer_1999}, as well as similarly UV-bright high-redshift sources (e.g.\ \citealt{Seibert_2002, Reddy_2008, Reddy_2012b, Pannella_2009, Heinis_2013, Bourne_2017, McLure_2018}). The local IRX-$\beta$ relation is, therefore, often used in order to attempt to correct for the dust absorption in high redshift ($z>2$) galaxies, when no IR constraints are available (e.g.\ \citealt{Treyer_2007, Gonzalez_2010, Gonzalez_2014, Bothwell_2011, Finkelstein_2012, Smit_2012, Smit_2016}).
More recent work has revealed that the local IRX-$\beta$ relation is not as tight as initially claimed, with significant scatter between individual galaxies (e.g.\ \citealt{Kong_2004}). It has been proposed that the inconsistencies may be explained with the introduction of a so-called `third parameter', examples of which include different ages of the underlying stellar populations (e.g.\ \citealt{Kong_2004, Burgarella_2005, Boquien_2009, Grasha_2013}), variations in the shapes of the intrinsic (without dust attenuation) UV slopes (e.g.\ \citealt{Boquien_2012}), and differences in dust types (e.g.\ \citealt{Thilker_2007}). Although the most extreme outliers may simply reflect a fundamental mismatch between the emission regions traced by the less-obscured UV and more-obscured FIR emission \citep{Goldader_2002, Trentham_1999}. More recently, it has been shown that the scatter in the IRX-$\beta$ relation may be driven by the diversity in the corresponding attenuation curves \citep{Salmon_2016, Salim_2019}.
Similarly, at high redshifts, {\it Herschel} and ALMA observations have shown that the most luminous FIR galaxies, when detected in the restframe UV, tend to have bluer UV slopes at fixed IRX (e.g.\ \citealt{Reddy_2010, Penner_2012, Oteo_2013, Casey_2014b, Watson_2015}). \citet{Casey_2014b} relates the deviation bluewards from the local IRX-$\beta$ relation to the IR luminosity, where they found that more luminous IR galaxies exhibit bluer UV colours. At the same time, it has been claimed that some high-redshift sources lie below the local relation \citep{Reddy_2010, Whitaker_2012, Capak_2015, Koprowski_2016, Bouwens_2016b, Pope_2017}, which has been suggested to be indicative that stellar light in these sources is affected by a different (e.g.\ SMC-like) dust extinction law.
Various theoretical efforts have attempted to explain the apparent deviations from the local IRX-$\beta$ relation \citep{Ferrara_2017, Safarzadeh_2017, Popping_2017, Narayanan_2018}. It was proposed that the most probable reason for the blue, dusty star-forming galaxies lying above the local relation, is the irregular relative covering of the dust and stars, where the least dust-attenuated regions produce blue UV colours, while the dusty parts give rise to the high values of the IRX. On the other hand, the galaxies with redder UV colours may be characteristic of older stellar populations, or affected by different dust extinction laws. Another possible explanation could be that the assumed dust temperatures for these sources are too low, leading to an underestimation of the IR luminosities and thus IRX.
In this work, we were able to precisely investigate the dust attenuation properties of high-redshift Lyman-break galaxies, as quantified via the IRX-$\beta$ relation, thanks to the availability of the FIR data. Matching $\sim$8000 3$\lesssim$$z$$\lesssim$5 LBGs with a sample of 708 submm-luminous sources, detected as a part of the ALMA survey of sub-millimeter galaxies from the SCUBA-2 Cosmology Legacy Survey (S2CLS, \citealt{Geach_2017}) map of the UKIDSS Ultra Deep Survey (UDS) field \citep{Stach_2019}, we identified an exquisite sub-sample of 41 ALMA-bright Lyman-break galaxies, for which we have the high-resolution dust continuum detections. Using rest-frame UV-to-FIR SED fitting, we are able to quantify the physical properties of the ALMA-detected LBGs, and we compare these to the more typical, FIR-faint high-redshift LBGs (using 209 LBGs in the ALMA coverage which are undetected in those maps).
In Section\,\ref{sec:data} we describe the data used in our analysis. Section\,\ref{sec:analysis} describes how the SED fitting was performed. It includes a quantitative description of the method used for varying the assumed shapes of the attenuation curves (Section\,\ref{sec:sed}), a summary of the derived physical properties, including stellar masses and UV \& IR luminosities (Section\,\ref{sec:props}), as well as the methodology for measuring $\beta$ and IRX (Section\,\ref{sec:irx1}). Section\,\ref{sec:disc} provides a discussion of the results, and our main conclusions are summarised in Section\,\ref{sec:summ}.
Throughout the paper we use the \citet{Chabrier_2003} stellar IMF with assumed flat cosmology with $\Omega_{\rm m} = 0.3$, $\Omega_\Lambda = 0.7$ and H$_0 = 70 $\,km\,s$^{-1}$\,Mpc$^{-1}$.
\section{Data}
\label{sec:data}
The starting point for our analysis is the catalogue of
716 sub-millimetre sources detected in the 850-${\rm \mu m}$ map of the UDS field from the SCUBA-2 Cosmology Legacy Survey \citep{Geach_2017}. These were subsequently followed up with ALMA 870\,${\rm \mu m}$ (Band-7) imaging between November 2013 and May 2017,
with an initial pilot study of 30 of the brightest SCUBA-2 sources in Cycle 1 by \cite{Simpson_2017}, see also \citet{Simpson_2015,Simpson_2015b}, and then subsequent observations to complete the full sample in Cycles 3, 4 and 5 by \citet{Stach_2019}, see also \citet{Stach_2018}. For the completed AS2UDS survey,
sources were identified by \citet{Stach_2019} using cleaned, 0.5 arcsec FWHM-tapered continuum maps, with average depths of $\sigma_{870}=0.25, 0.34, 0.23 $ and 0.09 mJy beam$^{-1}$ (Cycles 1, 3, 4 and 5, respectively). \citet{Stach_2019} catalogued 708 sub-millimetre galaxies (SMGs) brighter than $S_{870}\gtrsim 1$ mJy (4.3$\sigma$, or a false-positive rate of 2\%) in the full sample. For more details of the calibration, analysis and catalogue see \citet{Stach_2019}.
The detection limit of $S_{870}=1$ mJy at $z$$\simeq$3 corresponds to the infrared luminosity of $\sim$$10^{12}\,{\rm L_\odot}$. It is, therefore, comparable to the individual detections for SCUBA2 850\,${\rm \mu m}$ deep surveys (e.g. \citealt{Geach_2017}) and somewhat deeper than the {\it Herschel} studies at these redshifts (e.g., \citealt{Casey_2014}), but significantly shallower than the luminosities reached with stacking of the high-redshift LBGs in the FIR maps (e.g., \citealt{Coppin_2015, Koprowski_2018}), as well as some of the other LBGs individual ALMA detections (e.g., \citealt{Capak_2015, Koprowski_2016}).
Photometric redshifts and multi-wavelength properties for the SMGs in the AS2UDS sample are presented in \citet{Dudzeviciute_2019}, where the full SEDs across the rest-frame UV-radio photometry, were modelled using the high-redshift version of the {\sc magphys} code \citep*{Cunha_2008, Cunha_2015, Battisti_2019}. This analysis includes the sub-millimetre imaging from the {\it Herschel} \citep{Pilbratt_2010}, as provided by the public releases of the HerMES \citep{Oliver_2012} and PEP \citep{Lutz_2011} surveys undertaken with the SPIRE \citep{Griffin_2010} and PACS \citep{Poglitsch_2010} instruments. Due to the large beam sizes of the {\it Herschel}-SPIRE data, deblended maps were produced following the de-blending procedure detailed in \citet{Swinbank_2014}, with ALMA, MIPS 24\,${\rm \mu m}$ and radio catalogue sources used as priors for the locations of sources contributing to the PACS and SPIRE map flux. See \citet{Dudzeviciute_2019} for more details of the photometry and {\sc magphys} analysis.
We matched the SMGs from the ALMA Band-7 $>4.3\,\sigma$ catalogue with the UDS LBG sample from \citet{Koprowski_2018}. We start by updating the LBG selection using the new UV/optical/near-IR DR11 UDS catalogue (from the DR8 used previously). The primary benefit is the improved depth of 0.7 magnitudes in $K$-band (Almaini et al.\ in prep.; Hartley et al.\ in prep.). We identify galaxies at $z$$\simeq$3 using UGR, or BVR, filters \citep{Steidel_1996}, extending to higher redshifts by simply shifting the colour space to longer wavelengths, as described by \citet{Ouchi_2004}, finding 8494 3$\lesssim$$z$$\lesssim$5 LBGs. Since the parent optical catalogue is selected at $K$-band ($K<25.3$), our resulting LBG sample is mass complete to a limit of ${\rm log}(M_\ast/{\rm M_\odot})\gtrsim9.5$.
In order to select the optimal search radius for the cross-matching (minimizing the false detection rate, whilst maximizing the number of associations), we performed a suite of Monte Carlo simulations in the following manner. In each of 1,000 realizations, an artificial LBG catalogue was constructed by shifting all the individual sources in the original catalogue in random directions over a distance of 10\,arcsec. The average number of random ALMA-LBG associations can then be found as a function of search radius and compared with the unshifted map cross-matching to yield a ratio of the false association rate. In this way, we found a false association rate of $\sim 3\%$ for a search radius of 0.7\, arcsec, rising to $\sim 10\%$ at 1.5\, arcsec and $\sim 30\%$ at 2.2\, arcsec. Out of 250 3$\lesssim$$z$$\lesssim$5 LBGs in the UDS field which fall within the primary beam coverage of the AS2UDS ALMA maps, 41 were found to have ALMA Band-7 detections ($\geq 4.3\sigma$) within a 0.7 arcsec search radius (corresponding to $\sim$\,5\,kpc at $z\sim 3$--$5$), where only one of our optical counterparts is statistically expected to be a false association.
We note that our analysis makes use of the remaining $\sim 200$ ALMA-undetected LBGs for which the 870-$\mu$m maps provide limits on their FIR emission.
\begin{figure*}
\begin{center}
\includegraphics[scale=0.77]{histograms.pdf}
\end{center}
\caption{Histograms of main physical properties for FIR-bright LBGs, their FIR-faint analogues and the ALMA-detected SMGs that were not selected as LBGs, with the vertical lines representing mean values. The ordinate represents the probability density function, with the area under the curve equal unity. The left panel shows stellar masses, derived from the SED fits to the available UV-FIR photometry, assuming varying-slope attenuation curves, while for the SMGs stellar masses were adopted from \citet{Dudzeviciute_2019}. The middle panel depicts the distribution of UV luminosities found from best-fit SEDs, while the right panel shows values of IR luminosities. The blue vertical line in the right panel represents the stack of LBGs lying within the ALMA stamps that were not individually detected in the FIR luminosity of $L_{\rm IR}=6.6^{+2.1}_{-1.6} \times 10^{10} {\rm L}_\odot$ (see Section\,\ref{sec:props}). It can be seen that the ALMA-bright LBGs tend to have much higher stellar masses and IR luminosities, yet also slightly lower UV luminosities, on average, compared with the ALMA-undetected LBGs, which together suggest that the ALMA-bright LBGs suffer from correspondingly higher rest-frame UV dust obscuration on average. When comparing to the SMGs that have not been classified as LBGs, ALMA-bright LBGs seem to sample the full SMG distribution, with no obvious biases towards the high/low $M_\ast$/$L_{\rm IR}$ SMGs, although they represent only a modest fraction of the total high-redshift SMG population ($\simeq 10\%$). However, they have significantly higher UV luminosities, required for them to be selected as LBGs.}
\label{fig:hist}
\end{figure*}
\section{SED fitting}
\label{sec:analysis}
As noted above, \citet{Dudzeviciute_2019} have applied the {\sc magphys} energy-balance model to fit the full multi-wavelength SED of all 708 SMGs in AS2UDS. A very useful feature of energy-balance codes such as {\sc mapghys}, or {\sc cigale} \citep{Noll_2009,Serra_2011,Boquien_2019}, is that the energy absorbed by dust in the rest-frame UV/optical is equated with the energy emitted in the IR. Knowing the amount of the stellar energy reprocessed by dust ($L_{\rm IR}$), one can therefore determine the level of dust attenuation in the UV through near-IR and hence, given the sufficiently deep multi-wavelength photometry, find the underlying attenuation curve and the corresponding intrinsic (before dust attenuation) stellar SED. We note that in reality the distribution of stars and dust in a given galaxy may be very complex, with multiple irregular holes in the dust and with a fraction of stars potentially situated in front of the dust. When modelling the attenuation in such a galaxy with {\sc cigale}, one assumes the shape of the attenuation curve which is in effect the mass-weighted average of the attenuation curves affecting individual regions, where the relative distribution of dust and stars is assumed to be roughly uniform.
In our analysis we adopt the photometric redshifts for the 41 LBGs in our sample from \citet{Dudzeviciute_2019}, however, as we wish to investigate the the influence of the reddening law on the relationship between the UV and IR emission of these sources, we have chosen to refit the UV and IR parts of the SEDs using a code which allows us to easily vary the attenuation curve. Hence, we repeat the SED fitting using the energy-balance code {\sc cigale}\footnote{\url{http://cigale.lam.fr/}} \citep{Noll_2009,Serra_2011}, which enables us to explore varying the shape of the assumed attenuation curve. {\sc cigale} uses \citet{Bruzual_2003} stellar population templates, \citet{Chabrier_2003} stellar IMF and four values of stellar metallicity (0.004, 0.008, 0.02 and 0.05). A range of different star-formation histories (SFHs) was assumed, including both single- and double-burst, instantaneous, exponentially declining and continuous models (see \citealt{Koprowski_2018} for details) and the dust emission is modeled following the prescriptions of \citet{Casey_2012c}.
While we have chosen to use the photometric redshifts for the FIR-bright LBGs derived by
\citet{Dudzeviciute_2019}, for the ALMA-faint sources, in the absence of useful photometric constraints on their FIR emission, redshifts were determined using the {\sc eazy} template-fitting code.
For this we used 11-band photometric coverage of the UDS ($UBVRi z JHK$[3.6][4.5]), as described in \citet{Hartley_2013} and \citet{Mortlock_2013}. All of these photometric redshifts were calibrated using $\sim$\,7,000 spectroscopic redshifts, with the resulting dispersion in $dz/(1+z)$ of $\sigma=0.018$ for {\sc eazy} (further details will be provided in Hartley et al.\ in prep.) and $\sigma=0.08$ for {\sc magphys} \citep{Dudzeviciute_2019}.
\subsection{Attenuation curves}
\label{sec:sed}
In order to parametrise attenuation curves, {\sc cigale} adds two modifications to the standard Calzetti curve from \citet{Calzetti_2000}. The first modification is the alteration of the slope of the original Calzetti curve, normalised at $V$-band:
\begin{equation}\label{eq:att1}k_{\lambda} = k_{\lambda,{\rm Cal}}\times \left(\frac{\lambda}{550{\rm nm}}\right)^\delta,\end{equation}
\noindent where negative values of the relative slope, $\delta$, give steeper curves than that of Calzetti. The normalisation of the attenuation curve has been tied to $V$ band for historical reasons, where it has been shown that changes in $k_\lambda$ are associated with changes in optical extinction shortward of $\sim 600$\,nm, while extinction in the NIR remains relatively invariant \citep{Clayton_1988, Cardelli_1988}.
Since, by definition $k_\lambda$ is the attenuation normalised by its colour excess, $k_\lambda=A_\lambda/E(B-V)$, the modified curve from Equation\,\ref{eq:att1} has to be normalised with the modified version of the colour excess, hence:
\begin{multline}\label{eq:att2}k_{\lambda} = k_{\lambda,{\rm Cal}}\times \frac{E(B-V)_{\rm Cal}}{E(B-V)} \times \left(\frac{\lambda}{550{\rm nm}}\right)^\delta = \\
k_{\lambda,{\rm C}}\times \frac{R_V}{R_{V,{\rm Cal}}} \times \left(\frac{\lambda}{550{\rm nm}}\right)^\delta,\end{multline}
\noindent where $R_{V,{\rm Cal}}=A_V/E(B-V)_{\rm Cal}=4.05$ \citep{Calzetti_2000}. Using the above formulae one can find the dependence of the modified version of the total to selective extinction in the $V$-band, $R_V$, on the relative slope of the attenuation curve, $\delta$, to be:
\begin{equation}\label{eq:att3}R_V=\frac{R_{V,{\rm Cal}}}{(1+R_{V,{\rm Cal}})(440/550)^\delta-R_{V,{\rm Cal}}}.\end{equation}
\noindent The second modification is the addition of the UV bump:
\begin{equation}\label{eq:att4}k_\lambda = k_{\lambda,{\rm Cal}}\times \frac{R_V}{R_{V,{\rm Cal}}}\times \left(\frac{\lambda}{550{\rm nm}}\right)^\delta + D_\lambda,\end{equation}
\noindent where from \citet{Fitzpatrick_1986} the UV bump follows a Drude profile:
\begin{equation}\label{eq:att5}D_\lambda(B) = \frac{B\lambda^2\times (35{\rm nm})^2}{(\lambda^2-(217.5{\rm nm})^2)^2+\lambda^2(35{\rm nm})^2},\end{equation}
\noindent centred at 217.5\,nm (FWHM of 35\,nm), with $B$ being the amplitude of the bump.
Since we are interested in the intrinsic slopes of the attenuation curves, we do not want to normalise them by their colour excess. Instead, we decided to use the absolute attenuation in the $V$ band:
\begin{equation}\label{eq:att6}\frac{A_\lambda}{A_V}=\frac{k_\lambda}{k_V}=\frac{k_\lambda}{R_V},\end{equation}
\noindent with two free parameters: the relative slope of the attenuation curve, $\delta$, and the magnitude of the UV bump, $B$.
\subsection{Physical properties}
\label{sec:props}
We use the {\sc cigale} SED fitting procedure described above in order to derive basic physical properties for our sample: stellar masses, $M_\ast$, and UV \& IR luminosities. The best-fit values for the stellar masses are shown in Figure\,\ref{fig:hist} for the ALMA-detected (green) and the ALMA-undetected (blue) LBG samples. In addition, we show the AS2UDS SMGs that have not been classified as LBGs (red), where the physical properties were adopted from \citet{Dudzeviciute_2019}. We allow the attenuation curve to vary, where in Section\,\ref{sec:mass} we discuss the effect it has on the resulting values of $M_\ast$. The UV luminosity at rest-frame 1600\AA\, is defined here as $L_{\rm UV}\equiv \nu_{1600}L_{1600}$, where the luminosity density at 1600\AA, $L_{1600}$, was determined from the best-fit SED (Figure\,\ref{fig:hist}). As noted above, the photometric redshifts for the ALMA-bright sample were calculated from the rest-frame UV-radio photometry, using the high-redshift version of the {\sc magphys} code \citep*{Cunha_2008, Cunha_2015}, and are presented in \citet{Dudzeviciute_2019}.
To check for consistency, we compare the estimated stellar masses from our {\sc cigale} analysis to those derived using {\sc magphys} by \citet{Dudzeviciute_2019}. The mean mass for the 41 LBGs from {\sc cigale} is ${M_\ast= 5.3\pm 1.7 \times 10^{10} {\rm M_\odot}}$, consistent with the estimate from {\sc magphys} within the 1-sigma error bars: ${M_\ast= 6.4\pm 1.4 \times 10^{10} {\rm M_\odot}}$.
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{SED_stack.pdf}
\end{center}
\caption{Best-fitting SEDs for 209 ALMA-faint LBGs lying within the primary beam coverage of the AS2UDS ALMA maps. The rest-frame UV--NIR photometry are median values, while the FIR point represents the $3\sigma$ ALMA detection found by stacking the 209 LBGs in the ALMA maps ($S_{870}=65\pm 20\,{\rm \mu Jy}$). The resulting attenuation curve slope is $\delta=-0.10 \pm 0.25$. As explained in Section\,\ref{sec:irx2}, the relatively steep slope is consistent with the stacked population lying below the local IRX-$\beta$ relation.}
\label{fig:sedst}
\end{figure}
As explained earlier, the focus of this work is to understand the influence of the
attenuation curve on the IRX-$\beta$ relation of high-redshift LBGs. We are therefore
employing the {\sc cigale} code for the SED fitting. Producing accurate fits, however, requires a large number of SED models to be fit to the photometry, which is computationally very expensive. For this reason, we follow \citet{Salim_2018}, who instead of fitting the multi-wavelength FIR and the UV-NIR photometry simultaneously, modified {\sc cigale} to include the IR luminosity as a single constraint in the energy-balance calculation.
This approach is ideal for our analysis as we are not interested in details of the IR emission, such as characteristic dust temperature, instead our focus is on the rest-frame UV properties.
Adopting this simplified approach therefore allows us to model the stellar emission more accurately, by increasing a number of available SFHs, as well as allowing for a more detailed sampling of the dust attenuation laws.
In order to determine simple estimates of the IR luminosity for each of the ALMA-detected LBGs, we fit the dust emission SEDs, using 185 FIR SED templates compiled by \citet{Swinbank_2014}. These include local galaxy templates from \citet{Chary_2001}, \citet{Rieke_2009}, and \citet{Draine_2007} as well as high-redshift starburst galaxies from \citet{Ivison_2010} and \citet{Carilli_2011}, with a range of dust temperatures spanning 19--60 K. We find the best-fitting SEDs using a $\chi^2$ minimization approach and determine the $L_{\rm IR}$ by integrating between the rest-frame 8--1000\,${\rm \mu m}$ emission. The resulting values are depicted in Figure\,\ref{fig:hist}, with the mean value of
${L_{\rm IR}= 5.0^{+2.6}_{-1.5}\times 10^{12} {\rm L_\odot}}$ consistent with the {\sc magphys} derived estimates from \citet{Dudzeviciute_2019} of $L_{\rm IR}= 5.5^{+2.8}_{-1.6}\times 10^{12} {\rm L_\odot}$ (with errors representing the standard deviation), where the small difference can be attributed to the variation in the FIR libraries used.
For the ALMA-undetected sources we measure their average ALMA Band-7 flux by inverse-variance stacking in the ALMA maps. We acquire a $3.2\sigma$ detection of $S_{870}=65\pm 20\,{\rm \mu Jy}$, which is consistent with the ALMA-non-detections stack of \citet{Koprowski_2016}, but significantly lower than the $z \simeq 3$ stacks of \citet{Coppin_2015} and \citet{Koprowski_2018}. This latter discrepancy is due to two main factors. One is that in these works the LBG samples used for stacking also would have included any FIR-detected sources. The other is that our parent OIR catalogue is 0.7 mag deeper in $K$-band, which means that about a half of the ALMA-undetected sample of this work, would not have been selected using the shallower DR8 version. Since the faintest $K$-band sources are also expected to be the least massive and, therefore, the least dusty ones, our FIR stacks should be, and indeed are, fainter than those of \citet{Coppin_2015} and \citet{Koprowski_2018}. To translate the stacked flux to the IR luminosity, we adopt the $z \simeq 3$ best-fit FIR SED from \citet{Koprowski_2018}, and integrate the SED between rest-frame 8--1000\,${\rm \mu m}$ (Figure\,\ref{fig:sedst}).
In order to determine the errors we have performed a simple Monte Carlo simulation, where we have varied the redshifts and the photometry of each source by a random offset sampled from a Gaussian distribution with $1\sigma$ equal to the corresponding errors. The $1\sigma$ systematic errors on each of the physical parameters were then found by calculating the standard deviations from modeled sources. All the resulting values are summarised in Table\,\ref{tab:prop}. The results are depicted in Figure\,\ref{fig:hist}.
\begin{figure*}
\begin{center}
\includegraphics[scale=0.77]{seds_beta_single.pdf}
\end{center}
\caption{Rest-frame UV part of the best-fit SEDs for the FIR-bright LBGs studied in this work (see Section\,\ref{sec:props} for details), with the UV slope range enclosed by two vertical grey lines. We mark the `bad' rest-frame UV SED fits in red.}
\label{fig:sedsall}
\end{figure*}
\subsection{Values of IRX \& \boldmath${\beta}$}
\label{sec:irx1}
To calculate IRX, we simply divided $L_{\rm IR}$ by $L_{\rm UV}$, which were measured in the previous Section. The values of UV slopes, $\beta$, can be determined in various different ways (see \citealt{Rogers_2013} for a review). In this work we decided to fit the power-law slope to the best-fitting SED between the rest-frame 125 and 250\,nm. This is because we possess only $\sim 4$ photometry points on average within the rest-frame UV range (see Figure\,\ref{fig:sedsall}) and therefore the UV slopes calculated directly from the photometry, as oppose to the best-fit SEDs, have larger uncertainties. We note that \citet{Meurer_1999} and \citet{Calzetti_2000} did not find any presence of the UV bump in their samples. We, therefore, set the amplitude of the UV bump (Equation\,\ref{eq:att5}) to 0, since it falls within the UV slope range (central rest-frame wavelength of 217.5\,nm and FWHM of 35\,nm) and could potentially affect the resulting $\beta$, by artificially shifting it towards bluer colours. The errors were calculated as in previous section.
It is also crucial, for the reasons explained in Section\,\ref{sec:irx2}, to determine the intrinsic (before dust attenuation) shape of the stellar emission SED, with the corresponding intrinsic UV slope, $\beta_{\rm int}$. This is important for two reasons. One is that the exact shape of the intrinsic stellar SED is required in order to translate the observed SED (given the amount of energy re-emitted by dust, $L_{\rm IR}$), into the attenuation curve. The other is that $\beta_{\rm int}$ is roughly connected to the age of the underlying stellar population, which in turn affects the position of a given source on the IRX-$\beta$ plane (see Section\,\ref{sec:irx2}). Together with the attenuation curves, the intrinsic stellar SEDs, with the corresponding values of $\beta_{\rm int}$, are determined by performing SED fits to the available rest-frame UV-FIR photometry.
\begin{figure*}
\begin{center}
\includegraphics[scale=0.77]{IRX_beta_2.pdf}
\end{center}
\caption{Infrared excess, IRX, as a function of the UV slope, $\beta$. The FIR-bright LBGs (small circles for good $\beta$ fits and small stars for bad $\beta$ fits; see Figure\,\ref{fig:sedsall}) and the FIR-faint stack (large square) are colour-coded with the the intrinsic (before dust attenuation) UV slope, mass-weighted age and the relative attenuation curve slope, $\delta$, (top, middle and bottom panels, respectively). It can be seen in the inset plot of the bottom panel, that the distance along the ordinate of a given source from the average relation, $\Delta{\rm IRX}$, most obviously correlates with the relative attenuation curve slope, $\delta$, with the Spearman correlation of $0.89$ and the two-sided $p$-value of $\ll 0.001$, indicating significant correlation. The best-fit linear function gives $\Delta{\rm log(IRX)}=(2.79\pm 0.22)\times \delta + (0.48\pm 0.06)$. We, therefore, conclude that the scatter in the IRX-$\beta$ plane is driven mainly by the variations in the shapes of the underlying attenuation curves (see Section\,\ref{sec:irx2} for details).}
\label{fig:irxb}
\end{figure*}
It is important to note, that the attenuation curve slopes and the star-formation histories (with the resulting intrinsic UV slopes) are degenerate. One can in principle increase the amount of dust attenuation by either assuming younger stellar populations (bluer $\beta_{\rm int}$), or making the attenuation curve slope shallower. However, fixing the attenuation curve to that of Calzetti will produce SED fits which have statistically higher values of $\chi^2$ (worse fits) than when the attenuation laws are allowed to vary.
Energy-balance SED fitting codes, such as {\sc cigale}, equate the amount of energy absorbed by dust in the rest-frame UV/optical bands, with the amount of energy re-emitted in the IR. Since our sources are IR-bright by selection, the SED fitting procedure will determine a need for a large amount of dust attenuation. Since, on the other hand, our LBG sample colour selection requires (by definition) detections in the rest-frame UV, it is unsurprising that the LBGs will tend to have relatively blue observed UV slopes. Applying a large amount of dust attenuation to a blue galaxy will yield a correspondingly bluer intrinsic UV slope. However, there is a limit to how intrinsically blue a galaxy can be. This limit is set by the youngest population of stars produced with the \citet{Bruzual_2003} stellar population templates and \citet{Chabrier_2003} stellar IMF. For the stellar population age of 1\,Myr, the bluest UV slope possible within the models is about $-2.7$. Fixing the \citet{Calzetti_2000} attenuation curve will, therefore, often result in the observed rest-frame UV section of the best-fit SEDs being too red, yielding poor fits to the photometry (large values of $\chi^2$), indicating that one indeed needs to allow the attenuation curves to vary in order to produce correct SED fits, or accept that the UV and FIR parts of the SED are spatially decoupled.
\section{Results \& Discussion}
\label{sec:disc}
\subsection{IRX-\boldmath${\beta}$ scatter}
\label{sec:irx2}
The main observational parameter that drives the average IRX-$\beta$ relation (Figure\,\ref{fig:irxb}) is the observed UV slope, $\beta$. To first order, the more dust attenuation in the rest-frame UV, the redder the UV slope, and the larger the corresponding value of the IR luminosity, i.e.\ IRX. However, as can be seen in Figure\,\ref{fig:irxb}, individual FIR-detected sources tend to lie above and to the left of the average relation. In order to explain the apparent scatter, we investigate the variations in two additional SED parameters, intrinsic UV slope and the dust attenuation curve shape.
As can be seen in Figure\,\ref{fig:irxbt}, for a fixed attenuation curve, variations in the intrinsic UV slopes produce a scatter (red curves). This is because, at a given value of the IRX (i.e.\ given amount of the dust attenuation), different intrinsic UV slopes (value of $\beta$ for ${\rm IRX} \rightarrow 0$) will yield a variation in the measured $\beta$. Similarly, varying attenuation curve slopes will result in multiple values of $\beta$ for a given IRX (blue curves).
\subsubsection{Scatter from the intrinsic UV slopes}
\label{sec:scat1}
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{IRX_beta_theory.pdf}
\end{center}
\caption{The IRX-$\beta$ plane with the two main SED fitting parameters responsible for the apparent scatter shown in Figure\,\ref{fig:irxb}. While the average relation is driven by the amount of dust present in a given galaxy, the addition scatter can be introduced by either variations in the intrinsic UV slopes (red curves), or the differences in the slopes of the assumed attenuation curves (blue curve).}
\label{fig:irxbt}
\end{figure}
In the top panel of Figure\,\ref{fig:irxb} we present the scatter in the IRX-$\beta$ plane colour-coded with the rest-frame UV slope of the intrinsic (before dust attenuation) stellar emission spectrum, $\beta_{\rm int}$. It can be seen that ALMA-detected LBGs tend to have bluer $\beta_{\rm int}$ than their FIR-faint analogues. This indicates that the former sample must have a population of very young stars, responsible for the relatively blue UV slopes. However, it is known that the most massive galaxies tend to be older (e.g., \citealt{Thomas_2005}, \citealt{McDermid_2012} and \citealt{Pacifici_2013}). From Figure\,\ref{fig:hist}, it is clear that the ALMA-bright sources are much more massive than the ALMA-faint sample and so they are expected to harbour a significant population of old stars. The only way to model the SEDs with blue UV slopes and a large values of the IR luminosity is, therefore, to introduce the young burst of star formation, which is, at least in part, disconnected from the dust. The emission from these young stars will then be allowed to escape the galaxy mostly unreddened and will in effect produce relatively blue rest-frame UV SEDs. However, no clear correlation between the intrinsic UV slope and the offset between a given source and the local, average relation can be seen, which indicates that this is not the main parameter driving the scatter in the FIR-bright LBGs.
It has been proposed (e.g.\ \citealt{Kong_2004, Burgarella_2005, Boquien_2009, Grasha_2013}), that the IRX-$\beta$ scatter can be driven by the differences in the ages of the stellar populations. The stellar ages, however, affect the IRX values for a given observed $\beta$ through the effect they have on the intrinsic shape of the stellar emission SED, in particular the intrinsic UV slope, which, as shown in the top panel of Figure\,\ref{fig:irxb}, does not seem to drive the scatter above the average relation.
In the middle panel of Figure\,\ref{fig:irxb} we show the IRX and $\beta$ values for the ALMA-bright sample and the ALMA-faint stack, colour-coded with the mass-weighted age of the underlying stellar population. We assume a double-burst star formation history (SFH), where the resulting dependence of the star formation rate on time is:
\begin{equation}\label{eq:sfr} \Psi(t) \propto {\rm exp}(-t_1/\tau_1)+f_{\rm m}{\rm exp}(-t_2/\tau_2), \end{equation}
\noindent with the star forming timescale, $\tau$, for both bursts of the star formation and the mass fraction of the late burst population, $f_{\rm m}$, being kept as free parameters. This allows a large variation of the SFHs, where a number of scenarios are possible, including both single-burst and double-burst, instantaneous, exponentially declining and continuous (constant) SFHs. The effective age of a given stellar population is, therefore, the mass-weighted mean value of the late, main burst and the young burst of star formation. As can be seen, there is no clear correlation between the mass-weighted age and the offset between a given source and the local, average relation. The likely reason for this is that the rest-frame UV range along which $\beta$ is calculated, 125 to 250\,nm, is sensitive to the most massive, young stars present in the model, which should make the intrinsic UV slope sensitive mainly to the age of the young burst of star formation, while ignoring the old burst. However, since we allow a large variations of the SFH models, the intrinsic UV slope does not even correlate with the age of the young burst alone.
In Figure\,\ref{fig:seds} we show two extreme examples of the evolution of the young component of stellar SEDs with age. The top panel depicts the case of the instantaneous model, where all the stars have been formed during an instantaneous burst of star formation, while the bottom panel shows the continuous model, where stars are being formed at the constant rate. In the former model, young stars are not being replenished by the ongoing star formation, therefore the intrinsic UV slope reddens considerably quickly with age. In the latter case, the young stars are being constantly added to the population, causing the slope to redden more slowly. In addition, $\beta_{\rm int}$ is known to be sensitive to the metallicity (e.g., \citealt{Salim_2019}), with metal-poor galaxies having shallower intrinsic UV slopes. However, this effect becomes apparent only for a relatively old stellar populations and, since the mean age of the young burst in our sample, responsible for the slope of the rest-frame UV part of the stellar SED, is $25\pm 6$\,Myr, the metallicity is expected to have a negligible effect on the resulting values of the UV slope.
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{seds.pdf}
\end{center}
\caption{The evolution of the power-law slope of the rest-frame UV part of the stellar emission spectrum with the age of the stellar population for the case of the instantaneous and the continuous star formation (top and bottom panel, respectively). The grey vertical lines represent the wavelength range along which $\beta$ is calculated (125--250\,nm). It can be seen that the form of the correlation of the intrinsic UV slope with the age of the stellar population depends on the star formation histories, which is relevant to the source of the scatter in the IRX-$\beta$ plane, as explained in Section\,\ref{sec:scat1}.}
\label{fig:seds}
\end{figure}
\subsubsection{Scatter from the attenuation curves}
\label{sec:scatt}
\begin{figure*}
\begin{center}
\includegraphics[scale=0.77]{IRX_beta_C.pdf}
\end{center}
\caption{IRX-$\beta$ relation, colour-coded with the intrinsic UV slope, for the ALMA-bright LBGs of this work, for the case when the attenuation law has been fixed on that of Calzetti. The inset plot shows the best-fit SED for AS2UDS.0480.0, where it can be seen that it is impossible to find a good fit without relaxing the attenuation law. The large IR luminosities of ALMA-bright LBGs forces too much dust attenuation in the optical, artificially shifting most of the sample towards redder UV slopes.}
\label{fig:irxbc}
\end{figure*}
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{att_crvs2.pdf}
\end{center}
\caption{Attenuation curves found for each galaxy in the sample studied in this work, colour-coded with the relative slope of the attenuation curve, $\delta$, (Section\,\ref{sec:sed}). The black thick line shows the attenuation curve of \citet{Calzetti_2000}. The range that the slopes of the attenuation curves span is about 1, while the average error is 0.1. This means that the scatter seen in the plot is real rather than a result of SED modeling uncertainties.}
\label{fig:att}
\end{figure}
A very interesting feature of the bottom panel of Figure\,\ref{fig:irxb} is that the IRX-$\beta$ scatter appears to be very well correlated with the relative slope of the attenuation curve, $\delta$ (Equation\,\ref{eq:att1}). We note here that ALMA-bright LBGs are, by selection, bright in the rest-frame UV, as well as FIR, which places them above the local IRX-$\beta$ relation. The intrinsic UV slope ($\beta$ for ${\rm IRX}\rightarrow 0$) for the galaxies lying on the local relation is $\sim$$-2.1$. Since the bluest UV slope available in our models is $\sim$$-2.7$, one can in principle, as shown in Figure\,\ref{fig:sedsall}, expect some galaxies with a very young population of stars, to end up above the local relation. However, the scatter of the intrinsic UV colors due to the age and/or metallicity for star-forming galaxies is quite small compared to the observed range, so it has a much smaller effect on the IRX-$\beta$ scatter than the attenuation curve. As shown in Fig. 6 of \citet{Salim_2019}, even for $-8.5<{\rm log(sSFR)}<-10.5$ the range of intrinsic beta is only $-2.5$ to $-2.1$ (0.4), whereas true observed slope extends over a range that is $10\times$ larger. In addition, with fixed attenuation law one simply cannot find good SED fits for most of the ALMA-bright LBGs. We show this in the inset plot of Figure\,\ref{fig:irxbc} for the AS2UDS.0480.0. The best-fit intrinsic UV slope for that source has been found to already have the minimum available value of $-2.7$. The large IR luminosity forces so much dust attenuation in the optical, that the resulting best-fit observed stellar SED is way too red, ie. {\sc cigale} simply cannot find a decent fit with the fixed attenuation law. As the consequence, the scatter in the IRX-$\beta$ plane of Figure\,\ref{fig:irxbc} (with fixed attenuation law) is artificially shifted towards redder UV slopes, as compared with the top panel of Figure\,\ref{fig:irxb}. The only way to properly model ALMA-bright LBGs is to let the attenuation law vary, in which case the sources to the left of the local relation in the bottom panel of Figure\,\ref{fig:irxb} tend to have shallower attenuation curves. The inset plot of the bottom panel in Figure\,\ref{fig:irxb} shows the relation between the vertical offset of the IRX values for FIR-bright LBGs and the average relation, $\Delta{\rm log(IRX)}$, and the relative attenuation slope, $\delta$. It can be seen that the two quantities do correlate, with the Spearman correlation of $0.89$ and the two-sided $p$-value of $\ll 0.001$. The line in
the inset panel in Figure\,\ref{fig:irxb} shows the best-fit linear function:
\begin{equation}\label{eq:corr}\Delta{\rm log(IRX)}=(2.79\pm 0.22)\times \delta + (0.48\pm 0.06).\end{equation}
One of the reasons behind the variations in the attenuation curves is the type of dust (namely, the distribution of the dust grain sizes) present in a given galaxy, encoded in the form of the intrinsic extinction curve. The accurate derivation of the extinction curve relies on the precise knowledge of the intrinsic stellar emission spectrum shape, as well as on the assumption that the uniform screen of dust is situated between the source of stellar emission and the observer, neither of which may be correct. The differences in the type of dust are in effect incorporated into the resulting shape of the observed attenuation curve. Variations in the dust extinction found within the Milky Way and Small and Large Magellanic Clouds have been found to be of about two orders of magnitude at the rest-frame UV bands (e.g.\ \citealt{Gordon_2003}). Since the corresponding differences in our sample are of about six orders of magnitude (Figure\,\ref{fig:att}), it is unlikely that any possible variations in the intrinsic extinction curves will drive the massive diversity of the observed attenuation curves plotted in Figure\,\ref{fig:att} alone.
\begin{figure*}
\begin{center}
\includegraphics[scale=0.3, trim=0cm 0cm 0cm 0cm, clip]{308.pdf}\includegraphics[scale=0.3, trim=0cm 0cm 0cm 0cm, clip]{266.pdf}\includegraphics[scale=0.3, trim=0cm 0cm 0cm 0cm, clip]{583.pdf}\\
\includegraphics[scale=0.3, trim=0cm 0cm 0cm 0cm, clip]{162.pdf}\includegraphics[scale=0.3, trim=0cm 0cm 0cm 0cm, clip]{6.pdf}\\
\end{center}
\caption{5\,arcsec $\times$ 5\,arcsec stamps of five FIR-bright LBGs covered by the high-resolution {\it HST} imaging \citep{Stach_2019}, with the WFC3/IR F160W image in the background and the contours representing the ALMA observations ($1\sigma, 2\sigma, ...$, and in the case of AS2UDS.0006.0: $1\sigma, 5\sigma, 10\sigma, 15\sigma$ and $20\sigma$). We plot the shape of the ALMA beam in the bottom left corner and quote the corresponding values of the relative slope of the attenuation curve. The stamps are placed in the order of the increasing value of the relative attenuation curve slope. If the shape of the attenuation law depended on the level of disturbance of the stars and dust alone, one would expect it to increase as we move from left to right. Since the background image shows the stellar emission longwards of the wavelength range along which the UV slope is calculated and we only possess five such detections, no definite statements can be made on the source of the variations in the shapes of the corresponding attenuation curves. Provided that relative morphology of stars and dust is predicted to produce large range of attenuation curve slopes \citep{Ferrara_2017, Safarzadeh_2017, Popping_2017, Narayanan_2018}, we conclude that a larger sample of FIR-detected LBGs with high-resolution optical imaging would be necessary to investigate this effect more quantitatively.}
\label{fig:stamps}
\end{figure*}
As mentioned above, the age of the underlying stellar populations can drive the IRX-$\beta$ scatter through the intrinsic UV slope. It can also, however, affect it by altering the shapes of the attenuation curves (Section\,\ref{sec:irx2}). Since the star formation is triggered in the dense clouds of the ISM, the young stars ($<10$\,Myr) will be embedded in a thick shell of gas and dust (e.g.\ \citealt{Wild_2011}). One can envisage how this might manifest itself in the steepening of the galaxy attenuation curve: the younger the stellar populations, the more young stars residing in their birth clouds, and the steeper the integrated attenuation curves. This effect was investigated by \citet{Narayanan_2018b}, who found that for stellar populations younger than $\sim 2 $\,Gyr, variations in stellar ages do not significantly affect the shapes of the corresponding attenuation curves. Since our sample resides at redshifts of $z\simeq3$, the differences in the ages of stellar populations will thus likely a have minor effect on the slopes of the attenuation curves in our sample.
Another property that affects the attenuation laws in galaxies is the relative morphology of stars and dust (e.g.\ \citealt{Ferrara_2017, Safarzadeh_2017, Popping_2017, Narayanan_2018}). The original IRX-$\beta$ relationship, parametrised in the work of \citet{Meurer_1999}, was calibrated on a sample of local compact starburst galaxies, where the dust is considerably well mixed with the stars and the resulting attenuation curve was found to be consistent with that of \citet{Calzetti_2000}. Although we note, that, once corrected for far-UV aperture losses \citep{Overzier_2011, Takeuchi_2012}, the resulting Meurer relation is actually shallower than what Calzetti law predicts for IRX-$\beta$ (e.g. \citealt{Salim_2018}). However, SMGs at high redshifts have often been found to have significantly disturbed morphologies (e.g.\ \citealt{Smail_2004,Chen_2015, Chen_2017, Hodge_2018, Gomez_2018}). In this scenario, a fraction of the stellar population will be disconnected from the dust, with a small fraction of the stellar light leaving the galaxy mostly unaffected by dust; and if this exceeds the obscured emission in the UV bands then this will produce apparently very blue UV colours. On the other hand, the IRX, driven mainly by the IR luminosity, will have large values, since these sources have been selected as FIR-bright. A similar effect was observed in \citet{Wang_2018}, where the slope of the attenuation curve was affected by the galaxy inclination.
In addition, as found from the estimated dust masses and derived median hydrogen column densities, using an assumed gas to dust ratio of $\delta_{\rm gdr}=90\pm 25$ \citep{Swinbank_2014}, the SMGs in the UDS field have, on average, a $V$-band obscuration of $A_V\sim 500$ mag towards the FIR-bright component \citep{Simpson_2017}. This is in contrast with the typical $A_V$ from the SED fitting for FIR-bright LBGs in this work of order $1$--$2$ mag, which represents the dust obscuration of the stellar continuum emission detectable shortward of $\sim 1\mu$m in the restframe, integrated over the whole extent of the galaxy.
The significant difference between these two reddening estimates suggests a real disconnect between the sites of on-going star formation (which are so highly obscured that they are optically thick into the FIR) and the more extended distribution of likely older stars which are detectable in the restframe UV and optical. While the presence of the former component can be inferred from the FIR
emission and so included in determining the estimated stellar sizes or mass profiles of
the system \citep{Lang_2019}, it cannot be directly probed in the restframe UV.
\begin{figure*}
\begin{center}
\includegraphics[scale=0.77]{IRX_mass_double.pdf}
\end{center}
\caption{IRX as a function of stellar mass, colour-coded with the slope of the attenuation curve, $\delta$ (top panel), and the specific star-formation rate, sSFR (bottom panel). The filled black circles are the stacking results for the combined high-redshift ($z=3$--$5$) LBG samples of \citet{Koprowski_2018}, fitted with a best-fit power-law. The large square shows the FIR-faint ALMA sample of this work. The grey dashed line depicts the rough selection limit for the AS2UDS sample, corresponding to the typical UV luminosity of $\sim 1\times10^{10}\,{\rm L_\odot}$ at the stellar mass of $\sim 1\times 10^{10}\,{\rm M_\odot}$, rising to $\sim 3\times10^{10}\,{\rm L_\odot}$ at $\sim 2\times10^{11}\,{\rm M_\odot}$ and the sensitivity limit of the AS2UDS of $1\times10^{12}\,{\rm L_\odot}$. It can be seen that, as opposed to the IRX-$\beta$ plane, the scatter is driven by the variations in the values of the sSFRs, with FIR-bright LBGs lying in the `starburst' regime, indicating that they could be undergoing a major merger or secular processes responsible for an enhanced phase of star-formation activity. This supports the scenario in which ALMA-detected LBGs have their shallow attenuation curves produced by a disturbed stellar morphology, likely the result of an enhanced period of vigorous star formation activity.}
\label{fig:irxm}
\end{figure*}
Hence it is expected, provided that the relative morphology of dust and stars is the driving factor of the IRX-$\beta$ scatter, that sources with flat attenuation curves (blue curves in Figure\,\ref{fig:att}), lying considerably above the local IRX-$\beta$ relation (blue points in the bottom panel of Figure\,\ref{fig:irxb}), would be expected to have, on average, relatively disturbed morphologies, while the galaxies below the local relation would be expected to have less disturbed disk-like structures. To investigate this we require high-resolution optical {\it HST} data, which we only possess for the small fraction of the UDS field covered by the {\it HST} Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey (CANDELS; \citealt{Grogin_2011, Koekemoer_2011}).
In Figure\,\ref{fig:stamps} we show the optical stamps from \citet{Stach_2019} for five IR-bright LBGs covered by the CANDELS data, where we show them in the order of increasing value of the relative slope of the attenuation curve, $\delta$, (decreasing slope of the attenuation curve, see Figure\,\ref{fig:att}). Ideally, we would like to use resolved imaging from the UV slope range of 125 to 250\,nm, in order to construct the resolved map of $\beta$, but since our sample is very dusty, they are undetected in the rest-frame UV {\it HST} imaging. Based on this small sample we can see no clear correlation between $\delta$ and the level of disturbance. However, other works seem to confirm that SMGs do tend to have disturbed optical morphologies (e.g.\ \citealt{Chen_2015, Chen_2017, Hodge_2018, Lang_2019}). Since {\it HST} WFC3/IR F160W imaging, at the redshift of our sample of $z\simeq 3$, samples the restframe near-UV it is sensitive to a mix of moderately obscured young and intermediate age stars, and we only possess five such detections, no significant statements can be made about the impact the relative morphology of stars and dust have on the shape of the resulting dust attenuation curves for FIR-bright LBGs. We conclude that a much larger sample of FIR-bright LBGs, with the high-resolution rest-frame UV to FIR data, is required in order to investigate this issue properly.
\subsection{IRX-\boldmath${M_\ast}$ scatter}
\label{sec:irxm}
It is often argued that the stellar mass should be used, instead of the UV slope, in order to calibrate a relationship for the IRX. This is motivated by the claims that the total stellar mass provides an indirect tracer of the amount of the dust attenuation (e.g.\ \citealt{Sobral_2012, Whitaker_2012, Heinis_2013, Alvarez_2016, Dunlop_2017, Reddy_2018, McLure_2018}). This is because, to first order, the more massive the galaxy means that more stars will have formed (higher SFR) and more dust will be present in the ISM (larger $L_{\rm IR}$), producing redder UV colours. In Figure\,\ref{fig:irxm} we plot the values of the IRX and stellar mass for high-redshift LBGs, colour-coded with the relative slope of the attenuation curve, $\delta$, and the specific star-formation rate, sSFR, with the average relation from \citet{Koprowski_2018} overplotted. It is clear that the FIR-bright LBGs, as in the case of the IRX-$\beta$, do not follow the average relation.
The total SFR is a sum of the dominant IR obscured SF and (negligible) observed unobscured UV SF contributions. In order to find the values for the SFRs, we follow \citet{Madau_2014}. For UV SFRs we multiply $L_{\rm UV}$ by $2.5\times 10^{-10} \,{\rm M_\odot\,yr^{-1}\,L_\odot^{-1}}$, while for IR SFRs, we multiply $L_{\rm IR}$ by $1.73\times 10^{-10} \,{\rm M_\odot\,yr^{-1}\,L_\odot^{-1}}$. Since these conversions assume \citet{Salpeter_1955} IMF, we also multiply both SFRs by a factor of 0.63 to convert to \citet{Chabrier_2003} IMF.
A clear scatter can be seen in Figure\,\ref{fig:irxm}, with the FIR-bright LBGs exhibiting significantly larger values of the IRX, given their stellar masses, than their FIR-faint analogues. Moreover, the scatter in the IRX-$M_\ast$ plane does not seem to be driven by the shape of the attenuation curves, as in the case of the IRX-$\beta$, but by the variations in the sSFRs. This is, of course, to be expected, since IRX-$M_\ast$ relationship is analogous to that between the SFR and stellar mass.
The relation between the stellar mass and the total SFR is quantified via the so-called star-formation `main sequence' (MS; e.g.\ \citealt{Noeske_2007, Michalowski_2012a, Speagle_2014, Koprowski_2016b, Michalowski_2017}). At a given redshift, normal star-forming galaxies form a broad trend between star-formation rate and stellar mass, whereas `starbursts' are offset towards higher sSFRs by a factor of $>$\,2--4 (e.g.\ \citealt{Daddi_2007, Elbaz_2011, Speagle_2014}, although see \citealt{Elbaz_2018}). For the FIR-bright LBGs, the relatively large values of the IRX, given their stellar masses, are well correlated with the sSFRs, which indicates that these sources are undergoing a burst of SF. Again, since we possess high-resolution optical imaging only for 5 of our sources, we conclude that a larger sample with high-res UV-FIR imaging is required in order to reach any quantitative conclusions.
\subsection{Fraction of obscured star formation}
\label{sec:fobs}
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{f_obscured_single.pdf}
\end{center}
\caption{The fraction of obscured star formation, $f_{\rm obs}\equiv {\rm SFR_{IR}/SFR_{UV+IR}}$, as a function of stellar mass. The solid line is the best-fit function found by \citet{Whitaker_2017}. The green open triangles in the bottom panel represent the stacking results for the combined $z\sim3$, $z\sim4$ and $z\sim5$ LBG samples from \citet{Koprowski_2018}, the red open square shows the FIR-faint ALMA sample of this work, while the open blue circles depict the FIR-bright LBGs. It can be seen that the SFRs for the FIR-bright LBGs are almost exclusively dominated by the dust component.}
\label{fig:fobs}
\end{figure}
We start by noting that the quartile redshift range spanned by the 41 ALMA-detected LBGs analysed here corresponds to $z\sim 2.7$--3.5, but that these LBGs comprise only $16\pm 2$\% (41 of 250) of the ALMA-detected population in AS2UDS in that redshift range from \citet{Dudzeviciute_2019}. Hence, this UV-bright subset is far from representative of the bulk of highly star-forming galaxies at these redshifts. Indeed, \citet{Dudzeviciute_2019} find that
$\sim 17$\% of the SMG population are undetected down to the $K$-band limit of $K=25.7$ of this study, most of which are expected to lie at $z\sim2$--4, demonstrating the incompleteness of even NIR-selected samples against the most active massive, dust-obscured galaxies at $z\gg2$.
As found by \citet{Whitaker_2017}, the fraction of obscured star formation, $f_{\rm obs}\equiv {\rm SFR_{IR}/SFR_{UV+IR}}$, is a strong function of the stellar mass, at least out to $z=2.5$, where most massive sources ($M_\ast> 10^{11}\,{\rm M_\odot}$) have their SFRs dominated by the dust component ($f_{\rm obs}\sim 1$). They also found that the relation does not undergo any redshift evolution. Since our sample lies at $z\simeq 3$, we decided to investigate how it compares with the lower-redshift sources of \citet{Whitaker_2017}.
We present our results in Figure\,\ref{fig:fobs}. The solid line shows the best-fit function of \citet{Whitaker_2017}. We show the stacking results for the $z\simeq 3$--$5$ LBGs sample of \citet{Koprowski_2018} and the stack of the ALMA-faint LBGs lying within the ALMA maps investigated in this work, as well as the ALMA-detected LBG sample. It is clear that both stacked LBG samples are consistent with the findings of \citet{Whitaker_2017}. This confirms that, on average, UV/optically-selected star-forming dropout galaxies have a strong relation between the fraction of obscured star formation and the stellar mass, which continues out to $z\sim3$ and beyond. However, the ALMA-bright LBGs do not follow this trend. As in the case of IRX-$M_\ast$ relation, the FIR-bright LBGs, having their SFRs dominated by the IR-luminous dust component ($f_{\rm obs}\gtrsim 0.95$), lie well above the average curve (which the more typical FIR-faint LBGs tend to follow).
\subsection{Stellar masses}
\label{sec:mass}
\begin{figure}
\begin{center}
\includegraphics[scale=0.77]{mass_comp.pdf}
\end{center}
\caption{Ratio of the stellar masses derived when allowing the attenuation curves to vary freely, to values determined assuming the Calzetti law (ordinate), as a function of the Calzetti-like stellar masses (abscissa), colour-coded with the relative slope of the attenuation curve, $\delta$, (see Figure\,\ref{fig:att}). It can be seen that, since fixing the attenuation curve on that of Calzetti tends to produce worse fits (larger values of $\chi^2$), assuming the Calzetti attenuation curve will often underestimate stellar masses for FIR-bright sources (see Section\,\ref{sec:mass} for details).}
\label{fig:mass}
\end{figure}
A very important consequence of the variation in the shapes of the attenuation curves is the fact that it affects the resulting values of the stellar mass. The procedure of determining stellar masses from the SED fitting relies on the translation of the observed best-fit SED into the intrinsic, dust-unaffected spectrum. The wavelengths which are driving stellar masses are the NIR wavelengths, since it is the old, evolved stars, that dominate the mass budget of a galaxy. The \citet{Calzetti_2000} attenuation curve is characteristic of the average population of LBGs (although see \citealt{Boquien_2012}, \citealt{Schaerer_2013} and \citealt{Reddy_2018}), while, as shown in Figure\,\ref{fig:irxb}, sources well above the local IRX-$\beta$ relation will have greyer (less steep) attenuation curves. For a fixed amount of energy re-emitted by dust in the IR, $L_{\rm IR}$, the amount of the stellar light attenuated by dust will be dominated by the most energetic rest-frame UV wavelengths. Altering the slope of the attenuation curve will have little effect on the dust obscuration at the rest-frame UV, since $L_{\rm IR}$ must be conserved. However, greyer curves will have lower ratio of UV-to-NIR relative attenuation and, since the dust obscuration at UV is rather insensitive to the slope of the attenuation law, shallower curves will produce higher intrinsic NIR fluxes, increasing the resulting stellar masses.
Physically, for a given observed UV--NIR photometry and fixed $L_{\rm IR}$, the age of the stellar population will increase with the shallowing of the attenuation curve, since shallower curves produce redder intrinsic UV slopes (e.g. \citealt{Leja_2019}). In this scenario more stars (i.e.\ higher stellar mass) will be required in order to produce a given IR luminosity. We show this in Figure\,\ref{fig:mass}, where the difference between the stellar masses calculated allowing the attenuation curves to vary and fixing them at the \citet{Calzetti_2000} shape is plotted as a function of the Calzetti-like stellar mass, colour-coded with the relative slope, $\delta$. Bearing in mind that fixing the attenuation curve on that of Calzetti tends to produce worse SED fits (larger values of $\chi^2$), it is clear that galaxies with the greyest curves (largest $\delta$) will tend to have their stellar masses underestimated by a factor 2$-$3$\times$, when Calzetti curve is incorrectly assumed. Since, as found by \citet{Casey_2014b}, sources well above the local IRX-$\beta$ relation, suffering from the greyest attenuation curves, have the largest values of the IR luminosity, this effect will be most prominent in the case of FIR-bright LBGs and SMGs.
\section{Summary}
\label{sec:summ}
We exploit the large AS2UDS ALMA survey of the UDS field to identify and study a sample of 41 rare sub-millimetre-detected LBGs. We have performed a detailed analysis, including SED fitting to the rest-frame UV--FIR photometry, in order to investigate their relation to the typical LBG samples at high-redshifts in terms of the IRX, UV slope and stellar mass. We summarise the main conclusions from our work as follows:
\begin{enumerate}
\item{We find that the FIR-bright LBGs are on average significantly more massive than their FIR-faint analogues, which supports the claimed correlation between the stellar mass and the SFR (e.g.\ \citealt{Speagle_2014}).}
\item{When performing the rest-frame UV-FIR SED fits to the available broad-band photometry, we found that the attenuation curves must be allowed to vary in order to produce reliable fits. With the individual values of the IRX and $\beta$ found from the best-fit SEDs for each source in our sample, we found that FIR-bright LBGs lie well above the average Calzetti-like IRX-$\beta$ relation, characteristic of more typical high-redshift LBGs. We confirm that, while the differences in the ages of the underlying stellar populations play a role, the scatter is mainly driven by the variations in the shapes of the attenuation curves, where the offset between a given galaxy and the average relation correlates tightly with the attenuation curve slope, with Spearman correlation of $0.89$ and the two-sided $p$-value of $\ll 0.001$. We find the relation to be of a linear form with $\Delta{\rm log(IRX)}=(2.79\pm 0.22)\times \delta + (0.48\pm 0.06)$.}
\item{We find the attenuation curves for FIR-bright LBGs to be varying by over six orders of magnitude in the rest-frame UV (see also \citealt{Dudzeviciute_2019}). While this in principle should mainly be driven by the level of relative disturbance of the stars and dust in a given galaxy (e.g. \citealt{Chen_2015,Chen_2017}), the differences in the dust type, manifested in the shapes of the intrinsic extinction curves, as well as different ages of the stellar populations (fraction of young stars residing in their birth clouds), cannot be ignored. We investigate how the shape of the attenuation curve may relate to the relative morphology of stars and dust using high-resolution {\it HST} data. However, with only five sources detected, we cannot find any evidence for a strong correlation and suggest that larger sample may be required.}
\item{We investigated the relationship between the IRX and $M_\ast$ for our sample and found that FIR-bright LBGs exhibit unusually large values of IRX, given their stellar mass, relative to the FIR-faint sources. In addition, we show that the scatter in the IRX-$M_\ast$ plane is driven by the variations in the sSFRs, where FIR-bright galaxies tend to exhibit large values, placing them in the `starburst' regime of the star-formation main sequence. This indicates that the LBGs detected in the FIR with ALMA are massive galaxies, with apparently enhanced star-formation rates, potentially driven by mergers (see also \citealt{Dudzeviciute_2019}). This supports a scenario in which the unusually shallow slopes of the attenuation curves, detected in these sources, are largely produced by the disturbed morphologies.}
\item{When stellar masses are to be determined from the SED fits, caution must be taken when assuming the shapes of the underlying attenuation curves. As found in this work, since the IR-brightest galaxies lie well above the local IRX-$\beta$ relation, then attenuation curves much greyer than that of Calzetti must be assumed, in order not to underestimate their stellar masses.}
\end{enumerate}
The FIR-bright LBGs are a SFR-incomplete sub-sample of LBGs, which by selection have both blue UV slopes and high IR luminosities, placing them above the local IRX-$\beta$ relation. We found that assuming relatively flatter attenuation curves is required in order to produce good SED fits, where the slope of the attenuation law correlates with the scatter in the IRX-$\beta$ plane. It is, therefore, incorrect to infer the IR luminosity from the slope of the rest-frame UV portion of the stellar SED, using the local IRX-$\beta$ relation, at least for this small sub-sample of high-redshift galaxies. We found that the shallow slopes of the attenuation curves for FIR-bright LBGs may be produced, at least in part, by the irregular morphology of stars and dust. However, due to the lack of high-resolution UV and IR data, we cannot reach any quantitative statements. We conclude that a large, statistically significant sample of high-redshift LBGs with high-resolution data is required in order to advance this job further.
\section*{Acknowledgements}
M.P.K.\ and K.E.K.C.\ acknowledge support from the UK's Science and Technology Facilities Council (grant numbers ST/M001008/1 and ST/R000905/1). K.E.K.C. is also supported by a Royal Society Leverhulme Trust Senior Research Fellowship (SRF/R1/191013). A.K.\ and M.P.K.\ acknowledge support from the First TEAM grant of the Foundation for Polish Science No. POIR.04.04.00-00-5D21/18-00. A.K.\ also acknowledges support from the Polish National Science Center grants 2016/21/D/ST9/01098 and 2014/15/B/ST9/02111. J.E.G.\ is supported by a Royal Society URF. I.R.S., U.D., B.G., J.M.S.\ and A.M.S.\ also acknowledge support from STFC (ST/P000541/1). J.L.W. acknowledges support from an STFC Ernest Rutherford Fellowship (ST/P004784/2). M.J.M.~acknowledges the support of the National Science Centre, Poland through the SONATA BIS grant 2018/30/E/ST9/00208. Support for this work was provided by the Polish National Agency for Academic Exchange through the project InterAPS.
The ALMA data used in this paper were obtained under programs ADS/JAO.ALMA\#2012.1.00090.S, \#2015.1.01528.S, \#2016.1.00434.S and \#2017.1.01492.S. ALMA is a partnership of ESO (representing its member states), NSF (USA) and NINS (Japan), together with NRC (Canada) and NSC and ASIAA (Taiwan), in cooperation with the Republic of Chile. The Joint ALMA Observatory is operated by ESO, AUI/NRAO, and NAOJ.
The UKIDSS project is defined in \citet{Lawrence_2007}. Further details on the UDS can be found in Almaini et al. (in prep). UKIDSS uses the UKIRT Wide Field Camera (WFCAM; \citealt{Casali_2007}). The photometric system is described in \citet{Hewett_2006}, and the calibration is described in \citet{Hodgkin_2009}. The pipeline processing and science archive are described in Irwin et al (in prep) and \citet{Hambly_2008}.
This work is based on observations taken by the CANDELS Multi-Cycle Treasury Program with the NASA/ESA HST, which is operated by the Association of Universities for Research in Astronomy, Inc., under NASA contract NAS5-26555.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 4,802
|
Q: Predecir valores utilizando forecast.Arima y xreg Estoy intentando predecir valores usando forecast.Arima y los argumentos xtreg y h
Mis códigos en R son:
monthlylFatal <- ts(Spainmonthly$fatalities, start=c(1995,1), end= c(2013,12), frequency=12)
Spainpps <- ts(Spainmonthly$pps.2007, start=c(1995,1), end=c(2013,12), frequency=12)
Spainfinanc.cris <- ts(Spainmonthly$financ.cris.2008, start=c(1995,1), end=c(2013,12), frequency=12)
xreg <- cbind(Spainpps,Spainfinanc.cris)
arima1=Arima(logmonthlylFatal,order=c(0,1,1), seasonal=c(1,1,2), xreg = xreg)
forecast.arima1=forecast.Arima(arima1, xreg = xreg, h=84)
forecast.arima1
Me gustaría predecir valores desde Enero 2014 hasta Diciembre 2020 (h=84), pero como resultado obtengo predicciones hasta más del 2030. Por qué?
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 6,811
|
{"url":"http:\/\/mathhelpforum.com\/algebra\/168132-factorise-hard.html","text":"1. factorise, hard\n\nhi i need to know how to factorise this:\n\ni know if it was 2x^2 - 121 it would = (x+11) (x-11)\nso for this one would it be (2x +11) (2x-11) ???\n\nif not could you please tell me where im going wrong!\n\nThanks!\n\n2. Originally Posted by andyboy179\nhi i need to know how to factorise this:\n\ni know if it was 2x^2 - 121 it would = (x+11) (x-11)\nso for this one would it be (2x +11) (2x-11) ???\n\nif not could you please tell me where im going wrong!\n\nThanks!\nif it is x^2 - 121 = (x -11)(x + 11 )\nbut 2x^2 - 121 is not equal to (x- 11)(x+11)\n\nur solution for the question is rite 4x^2 - 121 = (2x - 11)(2x +11)\n\n3. In general, [tex]a^2- b^2= (a- b)(a+ b)\/[tex]\n\nActually, $2x^2- 121= (\\sqrt{2}x)^2- 11^2= (\\sqrt{2}x- 11)(\\sqrt{2}x+ 11)$, not $(x- 11)(x+ 11)$. I presume that was a mis-type.\n\n$4x^2= (2x)^2$ so, yes, $4x^2- 121= (2x- 11)(2x+ 11)$.","date":"2013-12-09 04:11:40","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 4, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.4858655631542206, \"perplexity\": 2539.3272402721263}, \"config\": {\"markdown_headings\": false, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2013-48\/segments\/1386163870408\/warc\/CC-MAIN-20131204133110-00029-ip-10-33-133-15.ec2.internal.warc.gz\"}"}
| null | null |
Karana gemmifera is a species of moth of the family Noctuidae. It is found in India and Taiwan.
References
Moths described in 1858
Hadeninae
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 3,122
|
\section{Introduction}
Coherence is ultimately the most distinctive feature of quantum systems.
Finding a proper measure of the coherence present at different times
scales in a quantum dynamical system is the first essential step for
assessing the role of quantum interference in natural and artificial
processes. This is of particular relevance for those quantum evolutions
in which information or energy are transformed and transferred in
order to achieve a given task with high efficiency. In this context
the relevant questions are: how much and what kind of coherence is
created vs destroyed by the dynamical evolution? How does coherence
determine/enhance the performance of the given process? These are
in general difficult questions and to be answered they require an
appropriate and sufficiently comprehensive framework. A general and
fundamental formalism to describe quantum interference is provided
by the decoherent histories (DH) approach to quantum mechanics. DH
have mainly found applications to foundational issues of quantum mechanics
such as the formulation of a consistent framework to describe closed
quantum systems, the emergence of classical mechanics from a quantum
substrate, the solution of quantum paradoxes, decoherence theory,
quantum probabilities \cite{Griffiths,Griffiths-Book,Omnes,GMannHartle,GellMannHartle,ZurekHist,Hartle_NegProb2008,GMann-Hartle_2014-1}.
However, DH can also be a systematic tool for quantifying interference
in quantum processes, and discussing its relevance therein. Indeed,
DH provide a precise mathematical formalization of interference by
means of the the so called\emph{ decoherence matrix} $\mathcal{D}$. The latter
is built on the elementary notion of \emph{histories} and allows one
to describe the quantum features vs the classical ones in tems of
interference between \emph{histories}, or \emph{pathways} if one resorts
to the mental picture of the double slit experiments. It is however
difficult to quantify in a compact and meaningful way the content
of $\mathcal{D}$ and its implications for the dynamics of specific systems.
Our first main goal is therefore to define and test appropriate measures
allowing for the investigation of how interference can determine the
performance of a given quantum information processing task. Starting
from $\mathcal{D}$ and by its sub-blocks we define different functionals.
In particular, we introduce a global measure of coherence $\mathcal{C}$
able to describe the coherence content of a general quantum evolution
at its various time scales; an average (over different time-scales)
measure of coherence; and average mesure of interference between
histories leading to a specific output.\par
While the tools we introduce
are of general interest and application, in order to test them we
apply them to a specific but relevant instance of quantum dynamics
taking place in photosynthetic membranes of bacteria and plants: quantum
energy transport. Here the basic common mechanism is the following:
a quantum excitation is first captured by the system and then migrates
through a network of sites (chromophores) towards a target site, e.g.,
a reaction center, where the energy is transformed and used to trigger
further chemical reactions. There is now an emerging consensus that
efficient transport in natural and biologically-inspired artificial
light-harvesting systems builds on a finely tuned balance of quantum
coherence and decoherence caused by environmental noise \cite{Ishizaki-PCCP,Ishizaki-PNAS,Belcher-Virus,Kevin-Bardeen_Original,GiordaJCP},
a phenomenon known as environment-assisted quantum transport (ENAQT).
This paradigm has emerged with clarity in recent years, as modern spectroscopic
techniques first suggested that exciton transport within photosynthetic
complexes might be coherent over appreciable timescales~\cite{Engel}.
Indeed, a growing number of experiments has provided solid evidence
that coherent dynamics occurs even at room temperature for unusually
long timescales (of the order of $100fs$)~\cite{Collini,Panit}.
Efforts to describe these systems have led to general models of ENAQT~\cite{Kevin-Bardeen_Original,Mohseni,Mohseni2,Plenio,Cao},
depicting the complex interplay of three key factors: coherent motion,
i.e., quantum delocalization of the excitation over different sites,
environmental decoherence, and localization caused by a disordered
energy landscape. So far, the presence of coherence in light-harvesting
systems has been qualitatively associated to the observation of distinctive
`quantum features'. Originally, coherence was identified with `quantum
wavelike' behavior as reflected by quantum beats in the dynamics of
chromophore populations within a photosynthetic complex. Later works,
employing quantum-information concepts and techniques, have switched
attention towards quantum correlations between chromophores, in particular
quantum entanglement\cite{Caruso,Whaley,GiordaLHCII,Ishizaki}. Besides
being open to criticism (see, e.g.,~\cite{Miller,Tiersch}), these
approaches do not provide direct quantitative measures of coherence
in the presence of noise. Therefore, in what follows, we shall apply
the novel tools based on DH to a simple yet fundamental model of quantum
energy transfer. We will focus on a relevant trimeric subunit of the
Fenna-Matthews-Olson (FMO) complex, the first pigment-protein complex
to be structurally characterized \cite{Blankenship}. The trimer is
virtually the simplest paradigmatic model retaining the basic charcteristics
of a disordered transfer network and it can also be conceived as an
essential building block of larger networks. For simplicity, we will
use the well-known Haken-Strobl model~\cite{Haken} to describe the
interplay between Hamiltonian and dephasing dynamics. While the model
is an oversimplified description of the actual dynamics taking place
in real systems, it allows to spot out the essential features that
may determine the high efficiency of the transport. We shall initially
focus on a new coherence measure $\mathcal{C}$, based on the decoherence
matrix, and characterize its behavior verifying that it can consistently
identify the bases and timescales over which quantum coherent phenomena
are present during the evolution of the system. We shall then show how the
average coherence exhibited on those time scales can be connected
with the delocalization process. A more detailed analysis will be
aimed at distinguishing between constructive and destructive interference
affecting the histories ending at the site where the excitation exits
the photosynthetic structure. By using the decoherence functional,
we will show that the beneficial role of dephasing for the transport
efficiency lies in a selective suppression of destructive interference,
a fact that has been systematically suggested in the literature, but
never expressed within a general and comprehensive framework that
allows the quantitative evaluation of coherence and its effects.\par
The application of the introduced tools and methods based on DH to
a simple yet paradigmatic system shows how one can properly quantify
the coherence content of a complex quantum dynamics and elucidate the
role of coherence in determining the overall efficiency of the process~\cite{Pleniocoherence}.\par
The paper is structured as follows. In Section \ref{sec:Decoherent-histories}
we review the basic decoherent histories formalism. In Section \ref{sec:The-coherence-measure}
we define the measure of coherence $\mathcal{C}$ and describe its
meaning and properties. In Section \ref{sec:Trimer} we first introduce
the used model for describing the energy transport in the selected
trimeric complex. We then discuss the coherence properties of the
excitonic transport: By means of the appropriate measures based on
the decoherent histories formalism we identify the essential features
that may determine the high efficiency of the transport. In Section
\ref{sec:FMO} we briefly discuss how to extend our results to the
whole FMO complex. In Section \ref{sec:Conclusions} we summarize
our results an draw our conclusions.
\section{Decoherent histories\label{sec:Decoherent-histories}}
The formalism of decoherent (or consistent) histories was developed
in slightly different flavors by Griffiths~\cite{Griffiths,Griffiths-Book},
Gell-Mann~\cite{GMannHartle,GMann-Hartle_2014-1}, Hartle\cite{Hartle_NegProb2008}
and Omn\`es~\cite{Omnes}. DH provide a consistent formulation of quantum
mechanics where
probabilities of measurement outcomes are replaced by probabilities
of \textit{histories}. In this formulation, external measurement apparatuses
are not needed, and then one does not need to postulate a ``classical
domain\textquotedbl{} of observers. As a consequence, quantum mechanics
becomes a theory that allows the calculation of probabilities of sequences
of events within any closed system, including the whole universe,
without the necessity of invoking postulates about the role of measurement.
In this framework, the ``classical domain\textquotedbl{} can be seen
to emerge as the description of the system becomes more and more coarse-grained.\par
The idea of `histories' stems from Feynman's `sum-over-histories'
formulation of quantum mechanics. As is known, any amplitude $\langle\psi_{f}|U(t_{f}-t_{0})|\psi_{i}\rangle$
between an initial and a final state can be expressed as a sum over
paths, or histories: Upon inserting the identity decomposition $\mathbb{I}=\sum_{j}|j\rangle\langle j|=\sum_{j}P_{j}$
at differerent times $t_{1}\dots t_{N}$ we get
\begin{small}
\begin{align*}
\langle\psi_{f}|U(t_{f}-t_{0})|\psi_{i}\rangle= & \langle\psi_{f}|U(t_{f}-t_{N})\sum_{j_{N}}P_{j_{N}} U(t_{N}-t_{N-1}) \dots \\
\ & \dots U(t_{2}-t_{1})\sum_{j_{1}}P_{j_{1}}U(t_{1}-t_{0})|\psi_{i}\rangle= \\
= \sum_{j_{1\dots}j_{N}}\langle\psi_{f}|P_{j_{N}}(t_{N}) & \dots P_{j_{1}}(t_{1})|\psi_{i}\rangle
\end{align*}
\end{small}
where we use the Heisenberg notation $P_{j}(t)=U^{\dagger}(t-t_{0})P_{j}U(t-t_{0})$.
Thus the total amplitude $\langle\psi_{f}|U(t_{f}-t_{0})|\psi_{i}\rangle$
is decomposed as a sum of amplitudes, each one corresponding to a
different \textit{history} identified by a sequence of projectors
$P_{j_{N}}\dots P_{j_{1}}$. \par
The decoherent histories formalism assumes that histories are the
fundamental objects of quantum theory and gives a prescription to
attribute probabilities to (sets of) histories. A history is defined
as a sequence of projectors
at times $t_{1}<\dots< t_{N}$.
Probabilites can be assigned within exhaustive sets of exclusive histories,
i.e., sets of histories $\mathcal{S}_{N}=\{t_{1},\dots,t_{N},P_{j_{1}},\dots,P_{j_{N}}\}$
where subscripts $j_{1},\dots,j_{N}$ label different alternatives
at times $t_{1},\dots,t_{N}$. Histories are exhaustive and exclusive
in the sense that the projectors at each time
satisfy relations of orthogonality $P_{j}P_{k}=\delta_{jk}P_{j}$,
and completeness, $\sum_{j}P_{j}=\mathbb{I}$.
In other words, the projectors $P_{j}$ define a
projective measurement. Within a specified set, any history can be
identified with the sequence of alternatives $\mathbf{j}\equiv j_{1},\dots,j_{N}$
realized at times $t_{1},\dots,t_{N}$. \par
Different alternative histories can be grouped together with a procedure
called \textit{coarse-graining}. Starting from histories $\mathbf{j}$ and
$\mathbf{k}$ we can define a new, \textit{coarse-grained} history $\mathbf{m}=\mathbf{j}\lor\mathbf{k}$
by summing projectors for all times $t_{\ell}$ such that $j_{\ell}$
and $k_{\ell}$ differ:
\begin{align*}
P_{m_{\ell}}=P_{j_{\ell}}+P_{k_{\ell}} & \mbox{ if }j_{\ell}\neq k_{\ell}\\
P_{m_{\ell}}=P_{j_{\ell}} & \mbox{ if }j_{\ell}=k_{\ell},
\end{align*}
for all $\ell=1,\dots,N$. By iterating this procedure, one can obtain
more and more coarse-grained histories.
A special type of coarse-graining is the \textit{temporal coarse-graining}:
we group together histories $\mathbf{j},\mathbf{k},\dots,\mathbf{l}$ such that
such that at some time $t_{\ell}$ we have $P_{j_{\ell}}+ P_{k_{\ell}}+\dots P_{l_{\ell}}=\mathbb{I}$.
Then the coarse-grained history $\mathbf{m}=\mathbf{j}\lor\mathbf{k}\lor\dots\lor\mathbf{l}$
contains only one-projector (equal to the identity) at time $t_{\ell}$,
that can be neglected and hence removed from the string of projectors
defining the history. On the other hand \textit{temporal fine-graining}
can be implemented for example by allowing different alternatives
at a times $t_{k}\notin\{t_{1},..,t_{N}\}$. In particular, one can
create new sets of histories $\mathcal{S}_{N+1}=\{t_{1},\dots,t_{N},t_{N+1},P_{j_{1}},\dots,P_{j_{N}},P_{j_{N+1}}\}$
from a given one $\mathcal{S}_{N}$ by adding different alternatives
at time $t_{N+1}>t_{N}$; the sets $\mathcal{S}_{N+1}$ are fine grained versions
of the sets $\mathcal{S}_{N}$. \\
Once we specify the initial state $\varrho$ and the (unitary) time
evolution $U(t)$,
we can assign any history $\mathbf{j}$ a \textit{weight}
\[
w_{\mathbf{j}}=\mbox{Tr}[C_{\mathbf{j}}\varrho C_{\mathbf{j}}^{\dagger}],\quad\mbox{with}\quad C_{\mathbf{j}}=P_{j_{N}}(t_{N})\dots P_{j_{1}}(t_{1})
\]
where we use the Heisenberg notation $P_{j_{\ell}}(t_{\ell})=U(t_{\ell})P_{j_{\ell}}U(t_{\ell})^{\dagger}$.
When the initial state is pure, $\varrho=\ket{\psi_{i}}\bra{\psi_{i}}$
and the final projectors are one dimensional, $P_{j_{N}}=\ket{\psi_{j_{N}}}\bra{\psi_{j_{N}}}$,
this formula takes the simple form of a squared amplitude
\begin{equation}
w_{\mathbf{j}}=|\bra{\psi_{j_{N}}}P_{j_{N-1}}(t_{N-1})\dots P_{j_{1}}(t_{1})\ket{\psi_{i}}|^{2}.
\end{equation}
Weights cannot be interpreted as true probabilities, in general.
Indeed, due to quantum interference between histories, the $w_{j}$
do not behave as classical probabilities. Indeed, consider two exclusive
histories $\mathbf{j},\mathbf{k}\in\mathcal{S}$ and the relative coarse-grained
history $\mathbf{m}=\mathbf{j}\lor\mathbf{k}$ by: $P_{m_{\ell}}=P_{j_{\ell}}+P_{k_{\ell}},\forall\ell$.
If the $w_{\mathbf{j}}$ were real probabilities, we would expect $w_{\mathbf{m}}=w_{\mathbf{j}}+w_{\mathbf{k}}$.
Instead, what we find is
\[
w_{\mathbf{m}}=w_{\mathbf{j}}+w_{\mathbf{k}}+2Re(\mbox{Tr}[C_{\mathbf{j}}\varrho C_{\mathbf{k}}^{\dagger}]).
\]
Due to the non-classical term $Re(\mbox{Tr}[C_{\mathbf{j}}\varrho C_{\mathbf{k}}^{\dagger}])$,
representing quantum interference between the histories $\mathbf{j}$ and
$\mathbf{k}$, the classical probability-sum-rule is violated.
The matrix
\begin{widetext}
\begin{equation}
\mathcal{D}_{\mathbf{j}\mathbf{k}}=\mbox{Tr}[C_{\mathbf{j}}\varrho C_{\mathbf{k}}^{\dagger}]=\mbox{Tr}[P_{j_{N}}U(t_{N}-t_{N-1})\dots P_{j_{1}}U(t_{1})\varrho U(t_{1})^{\dagger}P_{k_{1}}\dots U(t_{N}-t_{N-1})^{\dagger}P_{k_{N}}]\label{Eq: decomatrix}
\end{equation}
\end{widetext}
is called \emph{decoherence functional} or \emph{decoherence matrix}.
The decoherence matrix can be thought of as a ``density
matrix over histories'': Its diagonal elements are the
weights of histories and its off-diagonal elements are interferences
between pairs of histories. The decoherence matrix has the following
properties: i) it is Hermitian ii) it is semipositive definite iii)
it is trace one iv) it is block-diagonal in the last index, $\mathcal{D}_{\mathbf{j}\mathbf{k}}=\delta_{j_{N}k_{N}}\mathcal{D}_{\mathbf{j}\mathbf{k}}$.
Weights of coarse-grained histories can be obtained by summing matrix
entries in an $n\times n$ block of the decoherence matrix corresponding
to the original fine-grained histories. For istance, the weight of
history $\mathbf{m}=\mathbf{j}\lor\mathbf{k}$ is obtained by summing entries of
a $2\times2$ block of the decoherence matrix:
\[
w_{\mathbf{m}}=\mathcal{D}_{\mathbf{j}\jj}+\mathcal{D}_{\mathbf{k}\kk}+\mathcal{D}_{\mathbf{j}\mathbf{k}}+\mathcal{D}_{\mathbf{k}\mathbf{j}}.
\]
A necessary and sufficient condition to guarantee that the probability
sum rule $w_{\mathbf{j}\lor\mathbf{k}}=w_{\mathbf{j}}+w_{\mathbf{k}}$ apply within a set of histories
is
\[
Re[\mathcal{D}_{\mathbf{j}\mathbf{k}}]=0,\forall\mathbf{j}\neq\mathbf{k}.
\]
This condition is termed as \emph{weak decoherence\cite{GellMannHartle}};
the necessary and sufficient condition that is typically satisfied
\emph{\cite{GellMannHartle}} and that we will adopt in the following
is the stronger one termed as \emph{medium decoherence}
\begin{equation}
\mathcal{D}_{\mathbf{j}\mathbf{k}}=0,\forall\mathbf{j}\neq\mathbf{k}.\label{Eq.: mediumdeco}
\end{equation}
Medium decoherence implies weak decoherence. Any exhaustive and set
of exclusive histories satisfying medium decoherence is called a \textit{decoherent
set}. The fundamental rule of DH approach is that probabilities can
be assigned within a decoherent set, each history being assigned a
probabability equal to its weight. If medium decoherence holds, the
diagonal elements of the decoherence matrix can be identified as real
probabilities for histories and we can write $\mathcal{D}_{\mathbf{j}\jj}=p_{\mathbf{j}}$.
\\
\\
Due to property iv), if we perform a temporal coarse-graining over
all times except the last, we obtain `histories' with only one projection,
$P_{j_{N}}$ at the final time $t_{N}$. These histories automatically
satisfy medium decoherence:
\begin{align*}
\sum_{j_{1},\dots,j_{N-1}}\sum_{k_{1},\dots k_{N-1}}\mathcal{D}_{\mathbf{j}\mathbf{k}} & = \\
= \delta_{j_{N}k_{N}}\mbox{Tr}[P_{j_{N}}(t_{N})\varrho P_{j_{N}}(t_{N})] & \equiv\delta_{j_{N}k_{N}}p_{j_{N}}
\end{align*}
where $p_{j_{N}}\equiv\mbox{Tr}[P_{j_{N}}(t_{N})\varrho P_{j_{N}}(t_{N})]$
is the probability that the system is in $j_{N}$ at time $t_{N}$.
Due to interference, the probability of being in $j_{N}$ at time
$t_{N}$ is \textit{not} simply the sum of probabilities of all alternative
paths leading to $j_{N}$, i.e, of all alternative histories with
final projection $P_{j_{N}}$. In formulas,
\[
p_{j_{N}}\neq\sum_{j_{1},\dots j_{N-1}}w_{\mathbf{j}}=\sum_{j_{1},\dots j_{N-1}}\mathcal{D}_{\mathbf{j}\jj}.
\]
The probability and the global interference of histories $\mathcal{I}_{j_{N}}(\tau)$
ending in $j_{N}$ can be thus expressed as
\begin{equation}
p_{j_{N}}(\tau)=\sum_{j_{1},\dots j_{N-1}}w_{\mathbf{j}}(\tau)+\mathcal{I}_{j_{N}}(\tau)\label{eq: interference and probability at ending site j}
\end{equation}
with $\tau=N\Delta t\,$. Destructive interference will happen when
$\mathcal{I}_{j_{N}}<0$, constructive interference when $\mathcal{I}_{j_{N}}>0$.
\par
The decoherent histories formalism is consistent with and encompasses
the model of environmentally induced decoherence\cite{ZurekHist}.
Given a factorization of the Hilbert space into a subsystem of interest
and the rest (environment), $\mathcal{H}=\mathcal{H}_{S}\otimes\mathcal{H}_{\mathcal{E}}$,
the events of a history
take the form $P_{j_{\ell}{j'}_{\ell}}=\tilde{P}_{j_{\ell}}\otimes\Pi_{{j'}_{\ell}}$
where $\tilde{P}_{j_{\ell}}$ and $\Pi_{j'_{\ell}}$ are projectors
onto Hilbert subspaces of $\mathcal{H}_{S}$ and $\mathcal{H}_{\mathcal{E}}$
respectively. Histories for $S$ alone can be obtained upon considering
appropriate coarse-grainings over the degrees of freedom of the environment,
such that the events are $\tilde{P}_{j_{\ell}}\otimes\mathbb{I}_{\mathcal{E}}$
where $\mathbb{I}_{\mathcal{E}}$ is the identity over $\mathcal{H}_{\mathcal{E}}$.
Upon introducing the time-evolution propagator $\mathcal{K}_{t\ t_{0}}$
as $\varrho(t)=U(t-t_{0})\varrho(t_{0})U(t-t_{0})^{\dagger}\equiv\mathcal{K}_{t\,t_{0}}[\varrho(t_{0})]$,
we can rewrite the decoherence matrix as:
\begin{align}
\mathcal{D}_{\mathbf{j}\mathbf{k}}= & \mbox{Tr}[\tilde{P}_{j_{N}}\mathcal{K}_{t_{N}\ t_{N-1}}[\tilde{P}_{j_{N-1}}\mathcal{K}_{t_{N-1}\ t_{N-2}}[\label{Eq: dmschrodinger}\\
& \dots\mathcal{K}_{t_{1}\ t_{0}}[\varrho_{0}]\dots]\tilde{P}_{k_{N-1}}]\tilde{P}_{k_{N}}]. \nonumber
\end{align}
If the initial state is factorized $\varrho(t_{0})=\tilde{\varrho}_{S}(t_{0})\otimes\varrho_{\mathcal{E}}(t_{0})$,
then the reduced density matrix $\varrho_{S}(t)=Tr_{\mathcal{E}}[\varrho(t)]$
evolves according to $\varrho_{S}(t)=\tilde{\mathcal{K}}_{t\ t_{0}}\varrho_{S}(t_{0})$
where $\tilde{\mathcal{K}}$ is the (non-unitary) reduced propagator\index{reduced propagator}
defined by
\begin{equation}
\begin{small}
\mbox{Tr}_{\mathcal{E}}[U(t-t_{0})\tilde{\varrho_{S}}(t_{0})\otimes\varrho_{\mathcal{E}}(t_{0})U^{\dagger}(t-t_{0})]=\tilde{\mathcal{K}}_{t\ t_{0}}[\tilde{\varrho}_S(t_{0})].
\end{small}
\end{equation}
If the evolution of the system and environment is Markovian,
we can write $\tilde{\mathcal{K}}_{t\ t'}=\tilde{\mathcal{K}}_{t-t'}$.
As proved by Zurek~\cite{ZurekHist}, under the assumption of Markovianity
we can rewrite the decoherence matrix in terms of reduced quantities
alone, i.e., quantities pertaining to the system only:
\begin{widetext}
\begin{align}
\mathcal{D}_{\mathbf{j}\mathbf{k}}= & \mbox{Tr}_{S}[\tilde{P}_{j_{N}}\tilde{\mathcal{K}}_{t_{N}\ t_{N-1}}[\tilde{P}_{j_{N-1}}\tilde{\mathcal{K}}_{t_{N-1}\ t_{N-2}}[\dots\tilde{\mathcal{K}}_{t_{1}\ t_{0}}[\varrho_{0}]\dots]\tilde{P}_{k_{N-1}}]\tilde{P}_{k_{N}}]. \label{Eq: reduceddec}
\end{align}
\end{widetext}
That is, the model of environmentally induced decoherence can be obtained
by applying the decoherent histories formalism to system and environment
together, and by coarse-graining over the degrees of freedom of the
environment.
\section{The coherence measure $C$\label{sec:The-coherence-measure}}
The DH approach provides the most fundamental framework in which the
transition from the quantum to the classical realm can be expressed. Indeed,
it is based on the most basic feature characterizing the quantum
world: interference and the resulting coherence of the dynamical evolution.
Despite being a well developed field of study, the DH history approach
lacks for a proper global measure of the coherence produced by the
dynamics at the different time scales. We therefore introduce a measure
that quantifies the global amount of coherence within a set of histories.
Assume projectors for all times $t_{\ell},\ell=1,\dots,N$ are taken
in a fixed basis
$\ket{e_{j}}$, $P_{j}^{\ell}=\ket{e_{j}}\bra{e_{j}}$.
Assume further that histories are composed by taking equally spaced
times between consecutive projections i.e., $t_{1}=\Delta t,\dots,t_{N}=N\Delta t$.
(in other words, histories correspond to projections applied in
the same basis and repeated at regular times). For such a set of histories,
consider the decoherence matrix
\[
\mathcal{D}_{\mathbf{j}\mathbf{k}}^{(N,P,\Delta t)}=\mbox{Tr}[C_{\mathbf{j}}^{(N,P,\Delta t)}\varrho C_{\mathbf{k}}^{(N,P,\Delta t)}]
\]
where $C_{\mathbf{j}}^{(N,P,\Delta t)}=P_{j_{N}}(N\Delta t)\dots P_{j_{1}}(\Delta t)$.
Take the von Neumann entropy of the decoherence matrix,
\begin{align}
h(P,N,\Delta t)=-\mbox{Tr}[\mathcal{D}^{(N,P,\Delta t)}\log\mathcal{D}^{(N,P,\Delta t)}].
\end{align}
Due to coherence between histories, $h_{N}$ differs from the `classical-like'
Shannon entropy of history weights
\begin{equation}
h^{(c)}(P,N,\Delta t)=-\sum_{\mathbf{j}}w_{\mathbf{j}}^{(N,P,\Delta t)}\log w_{\mathbf{j}}^{(N,P,\Delta t)}
\end{equation}
where $w_{\mathbf{j}}^{(N,P,\Delta t)}=\mbox{Tr}[C_{\mathbf{j}}^{(N,P,\Delta t)}\varrho C_{\mathbf{j}}^{(N,P,\Delta t)}]$
are the diagonal elements of $\mathcal{D}^{(N,P,\Delta t)}$, i.e., the weights.
The difference between the two quantities is wider if off-diagonal
elements of the decoherence matrix are bigger, i.e., if the set of
histories is more coherent. Let us define:
\begin{equation}
C(P,N,\Delta t)\equiv\frac{h^{(c)}(P,N,\Delta t)-h(P,N,\Delta t)}{h^{(c)}(P,N,\Delta t)}. \label{eq: Cfunction}
\end{equation}
We argue that $C(P,N,\Delta t)$ is suitable to be used as a general
measure of coherence within the set of histories defined by $P,N,\Delta t$.
Indeed, we can readily prove the following properties: \\
i) $0\leq C(P,N,\Delta t)<1$. $C(P,N,\Delta t)<1$ is obvious. To prove $C(P,N,\Delta t)>0$, let
us define a matrix $\tilde{\mathcal{D}}_{\mathbf{j}\mathbf{k}}^{(N,P,\Delta t)}=\delta_{\mathbf{j}\mathbf{k}}\mathcal{D}_{\mathbf{j}\mathbf{k}}^{(N,P,\Delta t)}$
where off-diagonal entries are set to zero. Since
\begin{align*}
\mbox{Tr}[\mathcal{D}^{(N)}\log\tilde{\mathcal{D}}^{(N)}] & =\sum_{\mathbf{j}}D_{\mathbf{j},\mathbf{j}}^{(N,P,\Delta t)}\log D_{\mathbf{j},\mathbf{j}}^{(N,P,\Delta t)}= \\
& = \mbox{Tr}[\tilde{\mathcal{D}}^{(N,P,\Delta t)}\log\tilde{\mathcal{D}}^{(N,P,\Delta t)}]
\end{align*}
we obtain that the numerator of (\ref{eq: Cfunction}) can be expressed
as a quantum relative entropy:
\begin{align*}
& \ h^{(c)}(P,N,\Delta t)-h(P,N,\Delta t)=\\
& -\mbox{Tr}[\tilde{\mathcal{D}}^{(N,P,\Delta t)}\log\tilde{\mathcal{D}}^{(N,P,\Delta t)}]+\mbox{Tr}[\mathcal{D}^{(N,P,\Delta t)}\log\mathcal{D}^{(N,P,\Delta t)}] \\
& =\mbox{Tr}[\tilde{\mathcal{D}}^{(N,P,\Delta t)}(\log\mathcal{D}^{(N,P,\Delta t)}-\log\tilde{\mathcal{D}}^{(N,P,\Delta t)})]\\
& =h(\mathcal{D}^{(N,P,\Delta t)}||\tilde{\mathcal{D}}^{(N,P,\Delta t)})\geq0
\end{align*}
where $h(A||B)\geq0$ is the relative entropy between $A$ and $B$. \\
ii) $C(P,N,\Delta t)=0$ iff $\mathcal{D}_{\mathbf{j},\mathbf{j}}^{(N,P,\Delta t)}=\tilde{\mathcal{D}}_{\mathbf{j},\mathbf{j}}^{(N,P,\Delta t)}$,
i.e., $C(P,N,\Delta t)$ vanishes if medium decoherence holds for
the set of histories, since the two quantities $h(N,P,\Delta t)$
and $h^{(c)}(N,P,\Delta t)$ coincide in this case. \par
Thus $C(P,N,\Delta t)$ is in essence a (statistical) distance between
the decoherence matrix $\mathcal{D}$ and the corresponding diagonal matrix
$\tilde{\mathcal{D}}^{(P,N,\Delta t)}$, renormalized so that its value lies
between $0$ and $1$. The greater are the off-diagonal elements of
$\mathcal{D}^{(P,N,\Delta t)}$, the greater the distance. The meaning of
$C(P,N,\Delta t)$ can be easily understood if we use the linear entropy,
a lower bound to the logarithmic version:
\begin{align*}
\ & h_{L}(P,N,\Delta t)=1-\mbox{Tr}[(\mathcal{D}^{(N,P,\Delta t)})^{2}],\\
\ & 1-h_{L}^{(c)}(P,N,\Delta t)=\mbox{Tr}[(\tilde{\mathcal{D}}^{(N,P,\Delta t)})^{2}].
\end{align*}
In this case, we obtain a `linear entropy' proxy of $C(N,P,\Delta t)$
as:
\begin{align*}
& C_{L}(P,N,\Delta t)\equiv\frac{h_{L}^{(c)}(P,N,\Delta t)-h_{L}(P,N,\Delta t)}{h_{L}^{(c)}(P,N,\Delta t)}=\\
& = \frac{\sum_{\mathbf{j}\neq\mathbf{k}}|\mathcal{D}_{\mathbf{j}\mathbf{k}}^{(N,P,\Delta t)})|^{2}}{1-\sum_{\mathbf{j}}|\mathcal{D}_{\mathbf{j}\jj}^{(N,P,\Delta t)})|^{2}}
\end{align*}
which is a simplified version that, by avoiding the diagonalization
of $\mathcal{D}^{(N,P,\Delta t)}$, helps containing the numerical complexity.
The measure introduced is well grounded on physical considerations.
In the following we will apply it to a simple system in order to check
its consistency, and later use it to characterize the coherence
properties of the evolutions induced by various regimes of interaction
with the environment. First, one has to check whether the measure properly
takes into account the action of the bath. In particular, if the bath
is characterized by a decoherence time $\gamma^{-1}$, it is known
(\cite{ZurekHist}) that on time scales $\Delta t\ge\gamma^{-1}$
the decoherence matrix becomes diagonal: The probability of a history
at time $t_{N+1}$ can be fully determined by its probability at time
$t_{N}$, since no interference can occur between different histories.
Indeed, the action of the bath is to create a\emph{ decoherent set
of histories} that are defined by a proper projection basis: the pointer
basis (\cite{ZurekHist}). Therefore, the fine-graining procedure obtained
by constructing a set of histories $\mathcal{S}_{N+1}$ via the addition
of a new complete set of projections in the same basis at time $t_{N+1}=(N+1)\Delta t$
to the set $\mathcal{S}_{N}$, should leave the coherence functional
$C$ invariant, i.e., $C(P,N+1,\Delta t)\approx C(P,N,\Delta t)$.
If instead $\Delta t<\gamma^{-1}$ the same fine-graining procedure
should lead to $C(P,N+1,\Delta t)\ge C(P,N,\Delta t)$.
Before passing to analyze a specific system, we want to focus on
the complexity of the evaluation of $\mathcal{D}$ and $\mathcal{{C}}$.
The dimension of the decoherent matrix grows with the dimension $d$
of the basis $P$ and the number $N$ of time instants that define
each history as $d^{2N}$. This exponential growth in principle limits
the application of the DH approach to small systems. However, as for
the system considered in this paper the computational effort is contained
due to the small number of subsystems (chromophores) and the small
dimension of the Hilbert space which is limited to the single-exciton
manifold. As we shall see, by limiting the choice of $N$ to a reasonable
number, the analysis can be fruitfully carried even on a laptop.
\section{Trimer\label{sec:Trimer}}
We now start to analyze decoherent histories in simple models of energy
transfer
comprising a small number $d$ of chromophores (sites). Neglecting
higher excitations, each site $i$ can be in its ground $\ket{0}_{i}$
or excited $\ket{1}_{i}$ state. We work in the single-excitation
manifold, and define the site basis as
\[
\ket{i}\equiv\ket{0}_{1}\dots\ket{1}_{i}\dots\ket{0}_{d}\qquad i=1\dots d,
\]
i.e., state $\ket{i}$ represents the exciton localized at site $i$.
On-site energies and couplings are represented by a Hamiltonian $H$
that is responsible for the unitary part of the dynamics. Interaction
with the environment is implemented by the Haken-Strobl model, that
has been extensively used in models of ENAQT~\cite{Mohseni,Plenio,GiordaLHCII,Cao}.
The effect of the environment is represented by a Markovian dephasing
in the site basis, expressed by Lindblad terms $L$ in the evolution,
as follows:
\begin{equation}
\dot{\varrho}=[H,\varrho]+\sum_{i}\gamma_{i}[2L_{i}\varrho L_{i}^{\dag}-L_{i}^{\dag}L_{i}\varrho-\varrho L_{i}^{\dag}L_{i}]\label{eq: HakenStrobl}
\end{equation}
where $L_{i}=|i\rangle\langle i|$ are projectors onto the site basis,
and $\gamma_{i}$ are the (local) dephasing rates. Furthermore, site
$d$ can be incoherently coupled to an exciton sink, represented by
a Linblad term
\[
k_{trap}[2L_{trap}\varrho L_{trap}^{\dag}-L_{trap}^{\dag}L_{trap}\varrho-\varrho L_{trap}^{\dag}L_{trap}]
\]
where $L_{trap}=|sink\rangle\langle e|$ and $k_{trap}$ is the trapping
rate. Contrary to other works, we neglect exciton recombination, as
it acts on much longer timescales $(\sim1\:ns)$ than dephasing and trapping.
\\
The global evolution is Markovian and can be represented by means
of the Liouville equation
\begin{equation}
\dot{\varrho}=\mathcal{L}(\varrho)=\mathcal{L}_{H}(\varrho)+\mathcal{L}_{\gamma}(\varrho)+\mathcal{L}_{{trap}}(\varrho)
\end{equation}
that can be simply solved by exponentiation,
\begin{equation}
\varrho(t)=e^{i\mathcal{L}t}(\varrho(0)).
\end{equation}
In the notation above, the propagator has the form $\tilde{\mathcal{K}}_{t't}=e^{i\mathcal{L}(t'-t)}$.
The efficiency of the transport can be evaluated as the leak of the
population $p_{e}(t)=\bra{e}\rho\ket{e}$ of the exit site $e$ towards
the sink:
\begin{equation}
\eta(t)=2k_{trap}\intop_{0}^{t}\bra{e}\rho\ket{e}.\label{eq: efficiency}
\end{equation}
The overall efficiency of the process is obtained by letting $t\rightarrow\infty.$
While its Markovianity limits the faithful description
of decoherence processes actually taking place in real photosynthetic
systems,\emph{ }the model retains the basic and commonly accepted
aspects of decoherence, that acts in the site basis: albeit in a complex
non-Markovian way, the protein enviroment measures the system locally
(i.e., on each site), thus\emph{ }destroying the coherence in the site basis
and creating it in the exciton basis. Note that the formalism can also
be applied to a `dressed' or polaronic basis where we include strong
interactions between chromophores and vibrational modes. That is,
to apply the DH method, one only needs a model in which an exciton
hops between sites, dressed or undressed. The model is therefore suitable
to readily implement the decoherent histories paradigm and to spot
the main basic features we are interested in and that are at the basis
of the success of ENAQT.
The FMO unit has $7$ chromophores and a complex energy and coupling
landscape with no symmetries. Energies and couplings (i.e., the Hamiltonian
$H$) can be obtained by different techniques: They can be extracted
by means of 2D spectroscopy as in~\cite{Cho} or computed through
ab initio calculations as in~\cite{Renger}, with similar but not
exactly equal results. This very complex struxture makes FMO far from
ideal as a first example to study. We thus prefer to start by working
with a much simpler, yet fully relevant subsystem: the trimeric
unit composed by the sites $1,2$ and $3$ of the FMO complex in the
notation of~\cite{Renger,Cho}). The first chromophore is the site
in which the energy transfer begins, while the third chromophore is
the site from which the excitation leaves the complex. The Hamiltonian
of the trimeric subunit is \cite{Renger}
\begin{equation}
H_{Renger}=\left(\begin{array}{ccc}
215 & -104.1 & 5.1\\
-104.1 & 220 & 32.6\\
5.1 & 32.6 & 0
\end{array}\right).\label{eq: HamTrimer Renger}
\end{equation}
The eigenenergies of the system ar given by\textbf{ $E_{+}=322.85\,cm^{-1},$$E_{-}=119.13\,cm^{-1},$
}$\,E_{3}=-6.98\,cm^{-1}$ which yields the eigenperiods
$T_{ij}=(2\pi\hbar/\Delta E_{ij}):$ $T_{+-}=0.163\,ps,\,T_{-3}=0.100\,ps,\,T_{+3}=0.264\,ps$.
Due its structure, the trimer is a chain composed by a pair of chromophores
($1,2$), degenerate in energy and forming a strongly coupled dimer,
and a third chromophore moderately coupled with the second one only.
Since in the following we suppose that the exciton starts from site
$1$, we expect a prominent role of the dimer in the dynamics, at least
in the first tens of femtoseconds.\\
In order to show how the DH analysis can be implemented,
in the following we are going to consider histories in the site and the energy bases,
with $N$ projections at times $n\Delta t,\:n=1,..,N$. We first
use the coherence function $C(P,N,\Delta t)$ introduced above (\ref{eq: Cfunction})
to evaluate the global coherence of the exciton transport process.
In order to test the behavior of $C$ for different values of dephasing,
in Fig \ref{Fig: Trimer Renger C function} we first plot $C$ as
a function of the time interval $\Delta t$ between projections for two values
of the dephasing rate: $i)\,\gamma=0$, corresponding to the full
quantum regime (Fig.\ref{Fig: Trimer Renger C function}) for the
site basis (a); $ii)\:\gamma=10$ corresponding to an intermediate
value of dephasing (Fig. \ref{Fig: Trimer Renger C function}) for
the site basis (b) and the energy basis (c).
\begin{figure*}[tbh]
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig1a.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig1b.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig1c.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig1d.ps}}\protect\caption{Coherence function $C(P,N,\Delta t)$ as a function of $\Delta t$
for the trimer with Hamiltonian \ref{eq: HamTrimer Renger} evaluated:
\textbf{(a)} in the site basis for $\gamma=0$, \textbf{(b)} in the site basis for $\gamma=10 ps^{-1}$ \textbf{(c)} in the exciton basis
for $\gamma=10 ps^{-1}$ \textbf{(d)} in the exciton basis for different values of dephasing $\gamma$. }
\label{Fig: Trimer Renger C function}
\end{figure*}
Before entering the discussion of the various regimes, we
note that as a function of the number of projections $N$ all curves
display the expected behavior: The increase (decrease) of the number
of projections corresponds to a temporal fine-graining (coarse-graining)
of the evolution; therefore, an increase (decrease) of $N$ should
imply an increase (decrease) of the amount of coherence between histories.
As shown in Fig. (\ref{Fig: Trimer Renger C function}) the function
$C$ correctly reproduces the fine (coarse) graining feature: the
qualitative behavior of $C$ as a function of $\Delta t$ is not
affected by the choice of $N$, while an increase of $N$ corresponds,
at fixed $\Delta t$, to an increase of $C$.
We will therefore use in the following the value $N=4$ that allows
for a neat description of the phenomena and for a reasonable computational
time.
As for the behavior at fixed $N$, we have that in the full quantum
regime ($\gamma=0$), the system obviously displays coherence in the
site basis only since
\[
\mbox{Tr}[\ket{E_{i}}\bra{E_{i}}e^{-iH\Delta t}\rho e^{iH\Delta t}\ket{E_{j}}\bra{E_{j}}]=\bra{E_{i}}\rho\ket{E_{i}}\delta_{i,j}
\]
$ $and the decoherence matrix $\mathcal{D}$ in the energy basis is diagonal
and independent on $\Delta t$ and $N$. This simply means that in
the full quantum regime histories in the exciton basis are fully decohered,\textbf{
}since the system is not able to create coherence among excitons.
Still in the full quantum regime, in the site basis, the coherence oscillates
as the exciton, starting at site $1$, goes back an forth along the
trimer, and the evolution builds up coherence in this basis, see Fig.\ref{Fig: Trimer Renger C function}
(a). In this regime, the trimer can be approximately seen as a dimer
composed by the first two chromophores, and the exciton performs
Rabi oscillations with a period given by $T_{+-}=160\,fs$; $C$
oscillates with half the period: for $\Delta t=80\,fs$ the exciton
is migrated mostly on site $2$ and $C$ has a minumum -- which is different
from zero since the exciton is partly delocalized on site $3$, and
the system therefore exhibits a non vanishing coherence.
For intermediate values of $\gamma\approx10\,ps^{-1}$ , Fig.\ref{Fig: Trimer Renger C function}
(b), the coherence in site basis as measured by $C$ correctly drops
down at $\Delta t\ge\gamma^{-1}$\cite{ZurekHist}. The dephasing
has a strong and obvious effect on the coherence between pathways:
coherence in this basis is a monotonically decreasing function of $\gamma$.
This is well highlighted by the global coherence function $C$, whose
maximal values are reduced by a factor of $\sim 3$ with respect to
those corresponding to full quantum regime. After a time $\tau_{decoh}=\gamma^{-1}$
the histories are fully decohered. Indeed, due to the specific model
of decoherence (\ref{eq: HakenStrobl}), which amounts to projective
measurements on $\ket{i}\bra{i}$ at each site with a rate $\gamma$,
the system kills the coherence in the site basis, which in turn corresponds
to the stable pointer basis for this model \cite{ZurekHist}, i.e., the
basis in which the density matrix is forced to be diagonal
by the specific decoherence model. On the other hand,
and for the same reason, the dynamics starts to build up coherence
in the exciton basis $\ket{E_{i}}\bra{E_{i}}$, see Fig. \ref{Fig: Trimer Renger C function}(c).
However, this coherence is later destroyed - on a time scale of approximately $0.2\:ps$ - since the
stationary state of the model is the identity. This effect is even more
evident if one compares the behavior of $C$ in the exciton basis for
different values of $\gamma$, as shown in Fig. \ref{Fig: Trimer Renger C function}(d):
$C$ grows with $\gamma$ and it lasts over longer time scales. This
feature is coherent with the expectations: the equilibrium state for
high $\gamma$ is the identity. Due to the projections implemented
by the environment in the site basis, the system is forced to create coherence
in the exciton basis. When $\gamma$ is very high a quantum Zeno effect
takes in, the dynamics is blocked, and the time required to reach
the equilibrium, and to destroy coherences in all bases, consequently grows.
This first analysis therefore shows that $C$ is indeed a good candidate
for assessing the global coherence properties of quantum evolutions.
For a fixed number of projections $N$, $C(\Delta t)$ can be interpreted
as a \emph{measure of the global coherence exhibited by the dynamics
over the time scale} $\Delta t$.
We now analyze in detail the specific features of quantum transport
for the trimer. The dynamics starts at site $1$ and evolves by
delocalizing the exciton on the other chromophores. In order to study
this process, we first use a measure of delocalization introduced in\cite{GiordaLHCII}
for the study of LHCII complex dynamics:
\begin{equation}
\mathcal{{H}}(t)=-\sum_{i}p_{i}(t)\ln p_{i}(t)
\end{equation}
that is simply the Shannon entropy of $p_{i}(t)$, the populations of the three
chromophores. This measure allows one to follow how much the exciton
gets delocalized over the trimer with time and in different dephasing
situations: $0\le\mathcal{{H}}(t)\le\ln(3)$, i.e., $\mathcal{{H}}$
is zero when the exciton is localized on a chromophore and it takes
its maximal value $\ln(3)$ when the population of the three sites are equal.
In Fig. \ref{Fig.: pop site 3 and delocalization} we plot both $\mathcal{{H}}(t)$
and the population $p_{3}(t)$ of site $3$ for different values of
$\gamma$. Due to the presence of interference, in the mainly quantum
regime ($\gamma=0.1,1\:ps^{-1}$), the exiton first delocalizes mainly
over the dimer and partly on the third site: The first maximum corresponds
to $t=40\:fs=1/4 \ T_{+-} $ when the system builds up a (close to
uniform) coherent superposition between sites $1$ and $2$, while
a non negligible part of the exciton is found in site $3$; indeed
$\mathcal{{H}}(t=40\,fs)\approx0.75>0.69$, the last value corresponding
to $\ln(2)$ i.e., to a uniform superposition over the sites $1$
and $2$ only. As the dynamics of the sytems extends to later times
we see that $\mathcal{{H}}(t)$ and $p_{3}(t)$
have an oscillatory behavior, whose main period is $1/2 \ T_{+-}$,
and which approximately corresponds to Rabi oscillations between site
$1$ and $2$, although the initial state fully localized in site
$1$ cannot be rebuilt due to the presence of site $3$. As for the
transport, we see that in this regime the system cannot take advantage
of the initial fast and high delocalization: the exciton bounces back
and forth over the trimer. In the intermediate regime $\gamma=16\:ps^{-1}$,
due, as we will later see, to the selective suppression of interference
processes, the initial speed up in delocalization is sustained by
the dynamical evolution, and the transfer rate to site $3$ is correspondingly
increased. For very high values of decoherence ($\gamma=100$) the
role of initial interference is suppressed and the initial speed-up
disappears: the environment measures the system in site basis at high
rates and the delocalization process is highly reduced.
\begin{figure}[t]
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig2a.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig2b.ps}}
\protect\caption{For the trimer (\ref{eq: HamTrimer Renger}): \textbf{(a)} Delocalization
$\mathcal{{H}}(t)$ and \textbf{(b)} $p_{3}(t)$ population of site $3$
as a function of time for different values of $\gamma=0.1,1,16,100$ $ps^{-1}$}
\label{Fig.: pop site 3 and delocalization}
\end{figure}
The optimal delocalization occurs in correspondence of $\gamma\approx16\:ps^{-1}$
and it can be interpolated with a double exponential function
\[
\mathcal{{H}}_{\gamma=16}(t)=c_{0}+c_{1}e^{-t/\tau_{1}}+c_{2}e^{-t/\tau_{2}}
\]
with $c_{0}=1.098=\ln3,\,c_{1}=-0.84,\,c_{2}=-0.373$. The first time
scale $\tau_{1}=23\:fs$ describes the initial fast quantum delocalization
process described above, while the second time scale $\tau_{2}=238\:fs$
the slower subsequent delocalization and the reaching of the equilibrium
situation, $\mathcal{{H}}(t=\infty)=\ln(3)$.
We now pass to systematically analyze the behavior of the coherence
of the evolution with respect to the strength of the interaction with
the environment and its relevance for the energy transport process.
As a first step we plot both $\mathcal{{H}}(\tau=N\Delta t)$ and
$\mathcal{{C}}(\Delta t)$ for different values of $\gamma$, Fig.
(\ref{Fig.: H and C various gamma}). The plots show that the coherence
function exhibits the required behavior: For small $\gamma=0.1$
, $\mathcal{{C}}(\Delta t)$ oscillates with period $1/2 \ T_{+-}$,
following the Rabi oscillations of the dimer. The minima occur at
$n/4 \ T_{+-}$ , showing that the exciton is ``partially'' localized
on site $1$ or $2$, and partially delocalized on site $3$. As $\gamma$
grows, the system becomes unable to create coherence on large time
scales; the decay of $\mathcal{{C}}(\Delta t)$ is mirrored by a the
reduction of the amplitude in the oscillations of $\mathcal{{H}}(\tau=N\Delta t)$.
\begin{figure}[t]
\includegraphics[width=0.4\textwidth]{fig3.ps}\protect\caption{Delocalization $\mathcal{{H}}(\tau=N\Delta t)$ and coherence function
$\mathcal{{C}}(\Delta t)$ for $\gamma=1,16,60$ $ps^{-1}$}
\label{Fig.: H and C various gamma}
\end{figure}
We now focus on the relevant time scales $\tau_{d}$ for the initial
fast delocalization process highlighted by our previous analysis,
which are of the order of tens to hundreds of femtoseconds. We
therefore introduce the following \emph{average measure of global
coherence of the evolution}
\[
Q_{\tau_{d}}(\gamma)=\frac{1}{\tau_{d}}\int_{0}^{\tau_{d}}C(\Delta t)\,d\Delta t.
\]
$Q_{\tau_{d}}(\gamma)$ is the average of the coherence exhibited
by the dynamics of the system at the time scales $\Delta t\in(0,\tau_{d})$.
In Fig. \ref{Fig.: Average C vs gamma} we show $Q_{\tau_{d}}(\gamma)$
for the trimer (\ref{eq: HamTrimer Renger}) in the site basis for different
values of $\tau_{d}$.\textbf{ }We first focus on the behavior of
$Q_{\tau_{d}}(\gamma)$ for values of dephasing in the range $\gamma\in(0,1)$ $ps^{-1}$.
In this range, for small time scales $\tau_{d}=20$ to $200\,fs$
the average global coherence $Q_{\tau_{d}}(\gamma)$ is approximately
constant and equals the value attained in the full quantum regime
i.e., $Q_{\tau_{d}}(\gamma)\approx Q_{\tau_{d}}(\gamma=0.1)$ . For
larger time scales ($\tau_{d}\approx1\:ps$) $Q_{\tau_{d}}(\gamma)$ rapidly decreases
with $\gamma$. This analysis shows that the behavior of $Q_{\tau_{d}}(\gamma)$
matches the expectations: the higher $\gamma$ the smaller the time
scales over which decoherence takes place, the lower the global coherence
of the dynamics. Along with $C(\Delta t)$ the functional $Q_{\tau_{d}}(\gamma)$
is therefore in general a good candidate for the evaluation of the
global coherence of open quantum systems evolution. As for the transport
dynamics, we focus on the timescale identified with the analysis of
$\mathcal{{H}}(t)$ for optimal dephasing; for $\tau_{d}=\tau_{1}=20\,ps$
and $\tau_{d}=40\,ps$ we see that the system indeed retains most
of the average coherence of the purely quantum regime up to the optimal
values of decoherence ($\gamma=16\:ps^{-1}$ in the figure), losing it afterwards; this is a clear indication that this phenomenon
is at the basis of the the fast initial delocalization process. Over
longer time scales, the relevance of coherence is highly suppressed\textbf{.}
\begin{figure}[t]
\includegraphics[width=0.4\textwidth]{fig4.ps}\protect\caption{Average coherence of the evolution $Q_{\tau_{d}}(\gamma)$ for the
trimer (\ref{eq: HamTrimer Renger}) for different values of $\tau_{d}$}
\label{Fig.: Average C vs gamma}
\end{figure}
We now deepen our analysis about the relevance of the coherence of
the evolution for the energy tranfer efficiency. To this aim we focus
on the basic feature that distingushes the classical and the quantum
regime: interference. In particular we focus on the sub-block $\mathcal{D}_{3}$
of the decoherence matrix $\mathcal{D}$ pertaining to the third chromophore,
which describes the set of histories in site basis ending at site
$3$. Due to interference the probability of occupation of the site
$3$ at time $\tau=N\Delta t$ can be written in terms of the the
histories ending at site $3$ $p_{3}(\tau)=w_{3}(\tau)+\mathcal{I}_{3}(\mathbf{\tau})$,
see (\ref{eq: interference and probability at ending site j}) .
\begin{figure}[t]
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig5a.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig5b.ps}}
\protect\caption{\textbf{(a)} interference $\mathcal{{I}}_{3}(\tau)$ of histories ending
in site $3$ for the trimer (\ref{eq: HamTrimer Renger}) as a function
of $\tau=N\Delta t$ for different values of $\gamma=0.1,1,10,100$;
\textbf{(b)} $\mathcal{{I}}_{3}(\tau)$ for different intermediate values
of $\gamma$}
\label{Fig.: Interference with gamma Trimer}
\end{figure}
In Fig. \ref{Fig.: Interference with gamma Trimer} we show $\mathcal{I}_{3}(\tau)$
for different values of dephasing. One has different regimes: for
$\gamma\gg1$, the set of histories in site basis is fully decohered;
$\mathcal{I}_{3}(\tau)\approx0$, the histories do not interfere with each
other and $p_{3}(\tau)\approx w_{3}(\tau)$ i.e, the probability is
simply the sum of diagonal elements of $\mathcal{D}_{3}$. In the mainly quantum
regime $\gamma\le1\,ps^{-1}$, $p_{3}(\tau)\neq w_{3}(\tau)$: after
the initial positive peak the histories interfere with each other,
globally the interference is mostly negative and therefore $p_{3}(\tau)\le w_{3}(\tau)$.
For intermediate values of decoherence $\gamma^{-1}\approx10\:ps$
the interference has a positive peak and then reduces to zero. While
the first initial fingersnap of positive interference that takes place
in the first $\approx80\:fs$ is common for all curves corresponding
to small and intermediate values of $\gamma$, the \emph{main effect
of the bath is displayed after this initial period of time: the decoherence
gradually suppresses interference, both the positive and the negative
one}; \emph{however, for intermediate values of $\gamma$ the effect
is stronger as for the negative part of the interference patterns}.
The environment thus implements what can be called a \emph{quantum
recoil avoiding effect:} it prevents the part of the exciton that --
thanks to constructive interference -- has delocalized on site $3$
to flow back to the the other sites.
\begin{figure}[t]
\includegraphics[width=0.4\textwidth]{fig6a.ps}
\includegraphics[width=0.4\textwidth]{fig6b.ps}
\protect\caption{For the trimer (\ref{eq: HamTrimer Renger}) \textbf{(a)} Average
positive (green), negative (blue), total (red) interference $\average{\mathcal{{I}}_{3}}$
of histories ending in site $3$ as a function of $\gamma$; \textbf{(b)}
total average interference $\average{\mathcal{{I}}_{i}^{Tot}}$ for
site $i=1$ (red), $i=2$ (blue), $i=3$ (green) $\gamma$.}
\label{Fig.: average interference site 3}
\end{figure}
In order to evaluate a possible advantage provided by the initial speed-up
in the delocalization process and by the interference phenomena
showed above one has to take into account another relevant time scale
of the transport process: the trapping time. Indeed, if the system
is to take advantage of the fast delocalization due to the coherent
behavior, the exit of the exciton should take place on time scales
of the order of the delocalization process. The theoretical and experimental
evidences show that this is the case: the trapping time $\tau_{trap}=k_{trap}^{-1}$
for the FMO complex is estimated in the literature to be of the order
of $0.2\:ps$ i.e., the exit of the exiton starts soon after the fast
delocalization due to quantum coherence has taken place. The role
of the interference between paths, in particular those leading to
site $3$, can therefore be appreciated by numerically evaluating
\begin{equation}
\average{\mathcal{I}_{i}^{\beta}}=\frac{{1}}{\tau_{trap}}\int_{0}^{\tau_{trap}}\mathcal{I}_{i}^{\beta}(\tau)d\tau
\end{equation}
i.e., the average over the trapping time scale of $\tau_{trap}=200\:fs$
of the total ($\beta=Tot$), negative ($\beta=-$) and positive ($\beta=+$)$ $
average interference between the histories ending in site $i$, with
$\average{\mathcal{I}_{i}^{Tot}}=\average{\mathcal{I}_{i}^{+}}+\average{\mathcal{I}_{i}^{-}}$.
In particular, in Fig. \ref{Fig.: average interference site 3}(a)
the different kinds of interference are plotted for histories terminating
at site $3$: on average, the negative interference highly reduces
the total interference for small values of decoherence strength; when
$\gamma\approx10\:ps^{-1}$, $\average{\mathcal{I}_{3}^{-}}$ vanishes, the
average total interference equals the positive one $\average{\mathcal{I}_{3}^{Tot}}=\average{\mathcal{I}_{3}^{+}}$,
and it is maximal for values of $\gamma$ comparable to those that
maximize $\mathcal{{H}}(t)$ ($\approx16\:ps^{-1}$). In Fig. \ref{Fig.: average interference site 3}(b),
we compare the behavior of $\average{\mathcal{I}_{i}^{Tot}}$ for
all sites. The results again suggest that decoherence acts on the
interference provided by the quantum engine in order to favor the
flow of the exciton towards the exit chromophore: the average positive
interference between histories ending at sites $2$ and $3$ grows
in modulus with $\gamma$ and attains a maximum for intermediate values
of decoherence; while the average negative interference between histories
ending at site 1 decreases and attains a minimum for intermediate values
of $\gamma$. The combined effect of decoherence and interference
thus helps depopulating site 1 and populating site 2 and 3.
We can now tackle one of the most relevant aspects of our discussion:
the net effect of the above described phenomena on the overall efficiency
of the transport. The latter can be fully appreciated by evaluating
the efficiency of the process (\ref{eq: efficiency}) and by recognizing
that, in the decoherent histories language, it can be expressed as:
\[
\eta(t)=2k_{trap}\intop_{0}^{t}p_{3}(\tau)d\tau=W_{3}(t)+I_{3}(t)
\]
where $\tau=N\Delta t$ and $W_{3}(t)=2k_{trap}\intop_{0}^{t}w_{3}(\tau)d\tau,\ I_{3}(t)=2k_{trap}\intop_{0}^{t}\mathcal{{I}}_{3}(\tau)d\tau$.
This split allows one to appriciate the role of interference for the
efficiency. In Fig. (\ref{Fig.: efficiency vs interference trimer})
$\eta$ is plotted for different values of dephasing. In agreement
with what discussed above, we have three regimes: for very small values
of $\gamma$ the overall efficiency is poor; this is due to the presence
of high negative interference that in average prevents the exciton
to migrate to the exit site. For large values of $\gamma$ the interference
processes are completely washed out and the system cannot take advantage
of the fast quantum delocalization. For intermediate (optimal) values
of $\gamma$ only the negative interference has been washed out: $\mathcal{{I}}_{3}(\tau)$
is positive, it acts on short time scales, and it provides on average
an enhancement of the global efficiency.
\begin{figure}[t]
\includegraphics[width=0.4\textwidth]{fig7.ps}
\protect\caption{For the trimer (\ref{eq: HamTrimer Renger}): transport efficiency
$\eta(t)$, integrated weight $W_{3}(t)=2k_{trap}\intop_{0}^{t}w_{3}(\tau)d\tau$ and
integrated interference $I_{3}(t)=2k_{trap}\intop_{0}^{t}\mathcal{{I}}_{3}(\tau)d\tau$ for pathways
ending at site $3$ for different values of $\gamma=0.1,16,60$ $ps^{-1}$ and
$k_{trap}=5\,ps^{-1}$}
\label{Fig.: efficiency vs interference trimer}
\end{figure}
These results, within the limits of the simple model of decoherence
taken into account, undoubtedly show for the first time that the so
called ENAQT phenomenon can well and properly be understood both qualitatively
and quantitatively within the decoherence histories approach, i.e.,
in terms of very the basic concepts of coherence and interference
between histories. The often recalled ``convergence'' of time scales
or ``Goldilocks'' effect (\cite{Goldilocks}) in biological quantum
transport systems seems therefore to be well rooted in the processes
discussed above: if decoherence is too small the system shows both
positive and negative interference (see Fig. \ref{Fig.: pop site 3 and delocalization}),
the delocalization has an ocillatory behavior, and the exciton bounces
back and forth along the network thus preventing its efficient extraction.
If instead decoherence is very high one has that the complete washing
out of intereference and coherence implies the delocalization process
to be very slow, no matter how fast the trapping mechanism try to
suck the exciton out of the system. In order to take advantage of
the effects of quantum coherent dynamics: $i)$ the bath must act
on the typical time scales of quantum evolution in order to implement
the quantum recoil avoiding process; $ii)$ the extraction of the
exciton from the complex, characterized by $k_{trap}$, must then
start soon after the initial fast delocalization has taken place.
Should the extraction take place on longer time scales, the benefits
of the fast initial delocalization would be spoiled: Waiting
long enough, the system would eventually reach together with equilibrium
a decent delocalization even for moderately high values of $\gamma$,
but in this case the transfer would be obviously much slower.
\section{FMO\label{sec:FMO}}
\begin{figure}[t]
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig8a.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig8b.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig8c.ps}}
\protect\caption{FMO: \textbf{(a)} $\mathcal{C}(\Delta t)$ for different values
of $\gamma$; \textbf{(b)} $\mathcal{I}_{3}(\Delta t)$ for different
values of $\gamma$; \textbf{(c)} $\mathcal{H}(\tau=N\Delta t),\mathcal{C}(\Delta t)$
for $\gamma=1,16,60\:ps^{-1}$;}
\label{Fig: FMO C various gammas and C vs H}
\end{figure}
\begin{figure}[t]
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig9a.ps}}
\subfigure[]{\includegraphics[width=0.4\textwidth]{fig9b.ps}}
\protect\caption{FMO: \textbf{(a)} average interference $<\mathcal{I}_{3}>$ for
$k_{trap}=5\,ps^{-1}$; \textbf{(b)} transport efficiency $\eta(t)$, integrated weight $W_{3}(t)$ and
integrated interference $I_{3}(t)$ for pathways
ending at site $3$ for different values of $\gamma=0,1,16$ $ps^{-1}$ and
$k_{trap}=5\,ps^{-1}$}
\label{Fig: FMO average I on 200 fs 3 site and efficiency}
\end{figure}
The above arguments can be easily applied to the whole FMO complex.
Fig.~\ref{Fig: FMO C various gammas and C vs H} and~\ref{Fig: FMO average I on 200 fs 3 site and efficiency}
show the application of the decoherent histories method to excitonic
transport in FMO. The main features of the behavior of $\mathcal{{C}},\mathcal{{H}},\mathcal{{Q}},\mathcal{{I}}_{3}$
and $\eta$ are maintained although obvious differences can be found
since the dynamics in now determined by the interplay of different
eigenperiods and interference paths are more complex. In particular,
Fig.\ref{Fig: FMO C various gammas and C vs H}(b), one can
observe a revival of positive interference $\mathcal{{I}}_{3}$ for
small values of $\gamma\simeq1$ , that does enhance the efficiency
for $t\approx1.5ps$, Fig. \ref{Fig: FMO average I on 200 fs 3 site and efficiency}(a);
but this is not sufficient to compensate the initial and subsequent
negative interference, thus impeding the reach of optimal values of
$\eta$. In general, compared to the trimer and as suggested by Fig.
\ref{Fig: FMO average I on 200 fs 3 site and efficiency}(a) the
maximum average positive coherence on short time scales is attained
for smaller values of $\gamma$. The overall picture is not significantly
affected if one decides to start the dynamics from site $6$ instead
of site $1$, as it often is reported in the literature.
\section{Conclusions\label{sec:Conclusions}}
The decoherent histories approach provides a general theory to study
the distinctive feature exhibited by quantum systems: coherence. However,
despite its generality and foundational character, in order to measure
the effects of coherence and decoherence the DH approach needs to
be complemented with a quantitative way to condense the information
contained in the basic object of the theory, i.e., the decoherence
matrix $\mathcal{D}$. In this paper we introduce a set of tools that
allow one to assess the (global) coherence properties of quantum (Markovian)
evolution and that can be used to relate the coherence content of
a general quantum dynamical process to the relevant figure of merits
of the given problem. We first define the \textit{coherence functional}
$C(P,N,\Delta t)$, that can be interpreted as a measure of the global
coherence exhibited by the dynamics in the basis $P$ over the time
scale\emph{ $\Delta t$.} While this measure is completely general,
one can further introduce other relevant tools tailored to the specific
system and type of system-environment interaction at hand. We thus
focus on a simple yet paradigmatic model of environmentally assisted
energy transfer where coherence effects have been shown to play a
significant role in determining the efficiency of the process: a trimeric
subunit of the Fenna-Matthews-Olson photosynthetic complex. Based
on $\mathcal{D}$ and $C(P,N,\Delta t)$ we define: $a)$ a measure
$\mathcal{Q}_{\tau}(\gamma)$ able to characterize the average coherence
exhibited by the dynamics of the system over the time scales $\Delta t\in(0,\tau)$
for a fixed value of the dephasing $\gamma$; $b)$ a measure of the
average interference $<\mathcal{I}_{i}>$ occurring between the histories
ending at a given ``site'' $i$. \\Within the specific model, we
first thoroughly assess the consistency of the behavior of $C(P,N,\Delta t)$
in the various regimes. We then show how the introduced tools allow
to study the intricate connections between the efficiency of the transport
process and the coherence properties of the dynamics. In particular
we show that the delocalization of the exciton over the chromophoric
subunit is strongly affected by the amount of (average) coherence
allowed by the interaction with the bath in the first tens to hundreds
of femtoseconds. If the system-bath interaction is too strong, coherence
is suppressed alongside the interference between different histories,
in particular those ending at the site where the excitation leaves
the complex. If the interaction is too weak the system exhibits high
values of coherence even on long time scales, but it also exhibits
negative interference between pathways ending at the exit site, a
manifestation of the fact that the exciton bounces back and forth
over the network thus preventing its efficient extraction. In the
intermediate regime i.e., when the different time scales of the system
(quantum oscillations, decoherence and trapping rate) converge, the
system shows high values of coherence on those time scales. The action
of the bath has a \textit{quantum recoil avoiding effect} on the dynamics
of the excitation: the \textit{benefits of the fast initial quantum
delocalization of the exciton over the network are preserved and sustained
in time by the dynamics}; in terms of pathways leading to the exit
site, the action is to s\textit{electively kill the negative interference
between pathways, while retaining the initial positive one.} These
effects can be explicitly connected to the overall efficiency of the
environment-assisted quantum transport: the gain in efficiency
for intermediate (optimal) values of decoherence can thus be traced
back to the basic concepts of coherence and interference between pathways
as expressed in the decoherent histories language. \\While the specific
decoherence model used (Haken-Strobl) is an oversimplified description
of the actual dynamics taking place in real systems, we believe
that our analysis allows to spot out the essential features that may
determine the high efficiency of the transport even in more complex
system-environment scenarios. \\In conclusion, the tools introduced
in this paper allow to thoroughly assess the coherence properties
of quantum evolutions and can be applied to a large variety of quantum
systems, the only limits being the restriction to Markovian dynamics
and the computational efforts required for high dimensional systems.
However, the extension to non-Markovian realms is indeed possible~\cite{Allegraprep},
and the use of parallel computing may allow the treatement of reasonably
large systems.
\begin{acknowledgements}
P.~Giorda and M.~Allegra would like to
thank Dr.~Giorgio ``Giorgione'' Villosio for his friendship, his
support, his always reinvigorating optimism, and his warm hospitality
at the Institute for Women and Religion - Turin, where this paper
was completed (\textit{cogitato, mus pusillus quam sit sapiens bestia,
aetatem qui non cubili uni umquam committit suam, quin, si unum obsideatur,
aliud iam perfugium elegerit}).\\
\ \\
P.~Giorda would like to thank Prof.~A.~Montorsi, Prof.~M.G.A.~Paris and Prof.~M.~Genovese for their kind help. \\
\ \\
S.~Lloyd would like to thank M.~Gell-Mann for helpful discussions.
\end{acknowledgements}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 7,219
|
Johan (Janne) Hollertz, född 6 maj 1832 i Stockholm, död 28 september 1891 i New York, var en svensk tecknare. och officer i den amerikanska nordstatsarmén.
Han var son till bankkassören Johan Hollertz och hans hustru Anna Maria. Efter studier i Uppsala arbetade han som tidningsman i Stockholm bland annat för Kapten Puff och Riksbladet. Efter en obehaglig historia i Stockholm 1859 rymde han till Amerika. Där deltog han i inbördeskriget för nordstatsarmén 1861-1865 och skall ha lämnat aktiv tjänst med kaptens grad. Han var sedan verksam som kontorist i New York fram till 1872 då han blev redaktör för den där samma år grundade tidningen Nordstjernan och 1875 blev han redaktör för tidningen Folkets röst i Jamestown. Han återvände till Nordstjernans redaktion 1888 där han förutom text försåg tidningen med illustrationer. I Sverige utgav han 1857 ett häfte med titeln Min skrifbok. Hollertz finns representerad vid bland annat Uppsala universitetsbibliotek.
Källor
Svenskt konstnärslexikon del III sid 167, Allhems Förlag, Malmö.
Noter
Svenska tecknare under 1800-talet
Födda 1832
Avlidna 1891
Män
Konstnärer från Stockholm
Svenskamerikaner
Personer som tjänstgjort i USA:s armé
Representerade vid Uppsala universitetsbibliotek
Svenska journalister under 1800-talet
Journalister från Stockholm
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 7,523
|
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html lang="en-GB" xml:lang="en-GB" xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8"/>
<title>Statistics of goeswith in UD_French-GSD</title>
<link rel="root" href=""/> <!-- for JS -->
<link rel="stylesheet" href="https://maxcdn.bootstrapcdn.com/font-awesome/4.4.0/css/font-awesome.min.css">
<link rel="stylesheet" type="text/css" href="../../css/jquery-ui-redmond.css"/>
<link rel="stylesheet" type="text/css" href="../../css/style.css"/>
<link rel="stylesheet" type="text/css" href="../../css/style-vis.css"/>
<link rel="stylesheet" type="text/css" href="../../css/hint.css"/>
<script type="text/javascript" src="../../lib/ext/head.load.min.js"></script>
<script src="https://cdnjs.cloudflare.com/ajax/libs/anchor-js/3.2.2/anchor.min.js"></script>
<script>document.addEventListener("DOMContentLoaded", function(event) {anchors.add();});</script>
<!-- Set up this custom Google search at https://cse.google.com/cse/business/settings?cx=001145188882102106025:dl1mehhcgbo -->
<!-- DZ 2021-01-22: I am temporarily hiding the search field to find out whether it slows down loading of the title page.
<script>
(function() {
var cx = '001145188882102106025:dl1mehhcgbo';
var gcse = document.createElement('script');
gcse.type = 'text/javascript';
gcse.async = true;
gcse.src = 'https://cse.google.com/cse.js?cx=' + cx;
var s = document.getElementsByTagName('script')[0];
s.parentNode.insertBefore(gcse, s);
})();
</script> -->
<!-- <link rel="shortcut icon" href="favicon.ico"/> -->
</head>
<body>
<div id="main" class="center">
<div id="hp-header">
<table width="100%"><tr><td width="50%">
<span class="header-text"><a href="http://universaldependencies.org/#language-">home</a></span>
<span class="header-text"><a href="https://github.com/universaldependencies/docs/edit/pages-source/treebanks/fr_gsd/fr_gsd-dep-goeswith.md" target="#">edit page</a></span>
<span class="header-text"><a href="https://github.com/universaldependencies/docs/issues">issue tracker</a></span>
</td><td>
<gcse:search></gcse:search>
</td></tr></table>
</div>
<hr/>
<div class="v2complete">
This page pertains to UD version 2.
</div>
<div id="content">
<noscript>
<div id="noscript">
It appears that you have Javascript disabled.
Please consider enabling Javascript for this page to see the visualizations.
</div>
</noscript>
<!-- The content may include scripts and styles, hence we must load the shared libraries before the content. -->
<script type="text/javascript">
console.time('loading libraries');
var root = '../../'; // filled in by jekyll
head.js(
// External libraries
// DZ: Copied from embedding.html. I don't know which one is needed for what, so I'm currently keeping them all.
root + 'lib/ext/jquery.min.js',
root + 'lib/ext/jquery.svg.min.js',
root + 'lib/ext/jquery.svgdom.min.js',
root + 'lib/ext/jquery.timeago.js',
root + 'lib/ext/jquery-ui.min.js',
root + 'lib/ext/waypoints.min.js',
root + 'lib/ext/jquery.address.min.js'
);
</script>
<h2 id="treebank-statistics-ud_french-gsd-relations-goeswith">Treebank Statistics: UD_French-GSD: Relations: <code class="language-plaintext highlighter-rouge">goeswith</code></h2>
<p>This relation is universal.</p>
<p>37 nodes (0%) are attached to their parents as <code class="language-plaintext highlighter-rouge">goeswith</code>.</p>
<p>37 instances of <code class="language-plaintext highlighter-rouge">goeswith</code> (100%) are left-to-right (parent precedes child).
Average distance between parent and child is 1.02702702702703.</p>
<p>The following 5 pairs of parts of speech are connected with <code class="language-plaintext highlighter-rouge">goeswith</code>: <tt><a href="fr_gsd-pos-NOUN.html">NOUN</a></tt>-<tt><a href="fr_gsd-pos-X.html">X</a></tt> (23; 62% instances), <tt><a href="fr_gsd-pos-ADV.html">ADV</a></tt>-<tt><a href="fr_gsd-pos-X.html">X</a></tt> (6; 16% instances), <tt><a href="fr_gsd-pos-ADJ.html">ADJ</a></tt>-<tt><a href="fr_gsd-pos-X.html">X</a></tt> (4; 11% instances), <tt><a href="fr_gsd-pos-PRON.html">PRON</a></tt>-<tt><a href="fr_gsd-pos-X.html">X</a></tt> (3; 8% instances), <tt><a href="fr_gsd-pos-ADP.html">ADP</a></tt>-<tt><a href="fr_gsd-pos-X.html">X</a></tt> (1; 3% instances).</p>
<pre><code class="language-conllu"># visual-style 4 bgColor:blue
# visual-style 4 fgColor:white
# visual-style 3 bgColor:blue
# visual-style 3 fgColor:white
# visual-style 3 4 goeswith color:blue
1 Pas pas ADV _ Polarity=Neg 3 advmod _ wordform=pas
2 de un DET _ _ 3 det _ _
3 sèche sèche-cheveux NOUN _ Gender=Masc|Number=Sing|Typo=Yes 0 root _ CorrectForm=sèche-cheveux
4 cheveux _ X _ _ 3 goeswith _ _
5 ni ni CCONJ _ _ 7 cc _ _
6 de un DET _ _ 7 det _ _
7 prise prise NOUN _ Gender=Fem|Number=Sing 3 conj _ _
8 rasoir rasoir NOUN _ Gender=Masc|Number=Sing 7 nmod _ _
9 dans dans ADP _ _ 11 case _ _
10 la le DET _ Definite=Def|Gender=Fem|Number=Sing|PronType=Art 11 det _ _
11 salle salle NOUN _ Gender=Fem|Number=Sing 7 nmod _ _
12 de de ADP _ _ 13 case _ _
13 bains bain NOUN _ Gender=Masc|Number=Plur 11 nmod _ SpaceAfter=No
14 . . PUNCT _ _ 3 punct _ _
</code></pre>
<pre><code class="language-conllu"># visual-style 2 bgColor:blue
# visual-style 2 fgColor:white
# visual-style 1 bgColor:blue
# visual-style 1 fgColor:white
# visual-style 1 2 goeswith color:blue
1 Peut peut-être ADV _ Typo=Yes 6 advmod _ CorrectForm=peut-être
2 être _ X _ _ 1 goeswith _ _
3 était être AUX _ Mood=Ind|Number=Sing|Person=3|Tense=Imp|VerbForm=Fin 6 cop _ SpaceAfter=No
4 -ce ce PRON _ Gender=Masc|Number=Sing|Person=3|PronType=Dem 6 nsubj _ wordform=ce
5 un un DET _ Definite=Ind|Gender=Masc|Number=Sing|PronType=Art 6 det _ _
6 loup loup NOUN _ Gender=Masc|Number=Sing 0 root _ _
7 ? ? PUNCT _ _ 6 punct _ _
</code></pre>
<pre><code class="language-conllu"># visual-style 8 bgColor:blue
# visual-style 8 fgColor:white
# visual-style 7 bgColor:blue
# visual-style 7 fgColor:white
# visual-style 7 8 goeswith color:blue
1 En en ADP _ _ 2 case _ wordform=en
2 1866 1866 NUM _ Number=Plur 11 obl:mod _ _
3 une un DET _ Definite=Ind|Gender=Fem|Number=Sing|PronType=Art 4 det _ _
4 catégorie catégorie NOUN _ Gender=Fem|Number=Sing 11 nsubj:pass _ _
5 nouvelle nouveau ADJ _ Gender=Fem|Number=Sing 4 amod _ SpaceAfter=No
6 , , PUNCT _ _ 7 punct _ _
7 grand grand-croix ADJ _ Gender=Masc|Number=Sing|Typo=Yes 4 appos _ CorrectForm=grand-croix
8 croix _ X _ _ 7 goeswith _ SpaceAfter=No
9 , , PUNCT _ _ 7 punct _ _
10 est être AUX _ Mood=Ind|Number=Sing|Person=3|Tense=Pres|VerbForm=Fin 11 aux:pass _ _
11 créée créer VERB _ Gender=Fem|Number=Sing|Tense=Past|VerbForm=Part 0 root _ SpaceAfter=No
12 . . PUNCT _ _ 11 punct _ _
</code></pre>
</div>
<!-- support for embedded visualizations -->
<script type="text/javascript">
var root = '../../'; // filled in by jekyll
head.js(
// We assume that external libraries such as jquery.min.js have already been loaded outside!
// (See _layouts/base.html.)
// brat helper modules
root + 'lib/brat/configuration.js',
root + 'lib/brat/util.js',
root + 'lib/brat/annotation_log.js',
root + 'lib/ext/webfont.js',
// brat modules
root + 'lib/brat/dispatcher.js',
root + 'lib/brat/url_monitor.js',
root + 'lib/brat/visualizer.js',
// embedding configuration
root + 'lib/local/config.js',
// project-specific collection data
root + 'lib/local/collections.js',
// Annodoc
root + 'lib/annodoc/annodoc.js',
// NOTE: non-local libraries
'https://spyysalo.github.io/conllu.js/conllu.js'
);
var webFontURLs = [
// root + 'static/fonts/Astloch-Bold.ttf',
root + 'static/fonts/PT_Sans-Caption-Web-Regular.ttf',
root + 'static/fonts/Liberation_Sans-Regular.ttf'
];
var setupTimeago = function() {
jQuery("time.timeago").timeago();
};
head.ready(function() {
setupTimeago();
// mark current collection (filled in by Jekyll)
Collections.listing['_current'] = '';
// perform all embedding and support functions
Annodoc.activate(Config.bratCollData, Collections.listing);
});
</script>
<!-- google analytics -->
<script>
(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){
(i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o),
m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m)
})(window,document,'script','//www.google-analytics.com/analytics.js','ga');
ga('create', 'UA-55233688-1', 'auto');
ga('send', 'pageview');
</script>
<div id="footer">
<p class="footer-text">© 2014–2021
<a href="http://universaldependencies.org/introduction.html#contributors" style="color:gray">Universal Dependencies contributors</a>.
Site powered by <a href="http://spyysalo.github.io/annodoc" style="color:gray">Annodoc</a> and <a href="http://brat.nlplab.org/" style="color:gray">brat</a></p>.
</div>
</div>
</body>
</html>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 2,319
|
Shun Hing Square är en skyskrapa i Shenzhen, Kina. Den är 384 meter hög och har 69 våningar. Den blev färdigbyggd år 1996 och var i ett år, 1996 till 1997, Kinas högsta byggnad tills Guangzhou's CITIC Plaza uppfördes. Den är nu rankad 4:a i Kina och 8:a i världen.
Skyskrapor i Shenzhen
WP:Projekt Kina
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 6,247
|
Arasia — род аранеоморфных пауков из подсемейства в семействе . Представители рода распространены в Австралии и на островах Папуа — Новой Гвинеи.
Виды
Arasia eucalypti Gardzinska, 1996 — Папуа — Новая Гвинея
Arasia mollicoma (L. Koch, 1880) — Квинсленд
Arasia mullion Zabka, 2002 — Новый Южный Уэльс
Ссылки
Salticidae: Diagnostic Drawings Library by Jerzy Proszynski 2003
Пауки-скакунчики
Паукообразные Австралии
Паукообразные Океании
Роды паукообразных
Таксоны, описанные Эженом Симоном
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 7,881
|
Q: I am using CSS flex and I want to remove the elements (hide them or remove them) when the elements wrap to the next line I have a html code with a lots of span elements and I am using CSS flex container to hold them, now I want them to be display:hidden (hide the elements) instead of wrapping in next row dynamically as the size of screen reduces.
Can anyone suggest me the solution.
Below is the html and css code I am using
* {
margin: 0;
padding: 0;
box-sizing: border-box;
}
body {
min-height: 100vh;
background: #0c192c;
}
.container {
position: relative;
width: 100%;
height: 100vh;
overflow: hidden;
}
.bubbles {
position: relative;
display: flex;
flex-wrap: wrap;
}
.bubbles span {
position: relative;
height: 1rem;
width: 1rem;
margin: 0 4px;
border-radius: 50%;
background: #4fc3dc;
box-shadow: 0 0 0 10px #4fc3dc44;
}
<div class="container">
<div class="bubbles">
<span style="--i:11"></span>
<span style="--i:12"></span>
<span style="--i:23"></span>
<span style="--i:18"></span>
<span style="--i:18"></span>
<span style="--i:21"></span>
<span style="--i:15"></span>
<span style="--i:13"></span>
<span style="--i:24"></span>
<span style="--i:10"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
<span style="--i:10"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
</div>
</div>
A: I can suggest these :
*
*You can change your flex-wrap: nowrap, so your content is not wrapping anymore
*You can set max-height and overflow-y together to prevent showing wrapped lines anymore
.bubbles {
max-height:1rem;
overflow-y:hidden;
}
A: Maybe you could change the height and width of the span elements to min-height and min-width
* {
margin: 0;
padding: 0;
box-sizing: border-box;
}
body {
min-height: 100vh;
background: #0c192c;
}
.container {
position: relative;
width: 100%;
height: 100vh;
overflow: hidden;
}
.bubbles {
display: flex;
}
.bubbles span {
min-height: 1rem;
min-width: 1rem;
margin: 0 4px;
border-radius: 50%;
background: #4fc3dc;
box-shadow: 0 0 0 10px #4fc3dc44;
}
<div class="container">
<div class="bubbles">
<span style="--i:11"></span>
<span style="--i:12"></span>
<span style="--i:23"></span>
<span style="--i:18"></span>
<span style="--i:18"></span>
<span style="--i:21"></span>
<span style="--i:15"></span>
<span style="--i:13"></span>
<span style="--i:24"></span>
<span style="--i:10"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
<span style="--i:10"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
<span style="--i:14"></span>
<span style="--i:26"></span>
<span style="--i:17"></span>
<span style="--i:22"></span>
<span style="--i:25"></span>
<span style="--i:24"></span>
</div>
</div>
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 9,219
|
/****************************************************************************
* Included Files
****************************************************************************/
#include <nuttx/config.h>
#include <stdint.h>
#include <time.h>
#include <assert.h>
#include <nuttx/arch.h>
#include <arch/irq.h>
#include <arch/board/board.h>
#include "up_arch.h"
#include "chip/a1x_timer.h"
/****************************************************************************
* Definitions
****************************************************************************/
/* Timer 0 will run at the rate of OSC24M with no division */
#define TMR0_CLOCK (24000000)
/* The desired timer interrupt frequency is provided by the definition
* CLK_TCK (see include/time.h). CLK_TCK defines the desired number of
* system clock ticks per second. That value is a user configurable setting
* that defaults to 100 (100 ticks per second = 10 MS interval).
*
* Timer 0 counts down from the interval reload value to zero, generating
* an interrupt (and reload) when the counts decrements to zero.
*/
#define TMR_INTERVAL ((TMR0_CLOCK + (CLK_TCK >> 1)) / CLK_TCK)
/****************************************************************************
* Private Types
****************************************************************************/
/****************************************************************************
* Private Function Prototypes
****************************************************************************/
/****************************************************************************
* Global Functions
****************************************************************************/
/****************************************************************************
* Function: up_timerisr
*
* Description:
* The timer ISR will perform a variety of services for various portions
* of the systems.
*
****************************************************************************/
int up_timerisr(int irq, uint32_t *regs)
{
/* Only a TIMER0 interrupt is expected here */
DEBUGASSERT((getreg32(A1X_TMR_IRQ_STA) & TMR_IRQ_TMR0) != 0);
/* Clear the pending interrupt by writing a '1' to the status register */
putreg32(TMR_IRQ_TMR0, A1X_TMR_IRQ_STA);
/* Process timer interrupt */
sched_process_timer();
return OK;
}
/****************************************************************************
* Function: up_timer_initialize
*
* Description:
* This function is called during start-up to initialize
* the timer interrupt.
*
****************************************************************************/
void up_timer_initialize(void)
{
uint32_t regval;
/* Set the timer reload interval value */
putreg32(TMR_INTERVAL, A1X_TMR0_INTV_VALUE);
/* Configure timer 0:
*
* ENABLE - Enable counting
* RELOAD - Reload timer interval value
* CLKSRC - OSC24M
* PRESCALER - No division
* MODE - Continuous mode
*/
regval = (TMR_CTRL_EN | TMR_CTRL_RELOAD | TMR_CTRL_SRC_OSC24M |
TMR_CTRL_CLK_PRES_DIV1 | TMR_CTRL_MODE_CONTINUOUS);
putreg32(regval, A1X_TMR0_CTRL);
/* Make sure that interrupts from the Timer 0 are disabled */
up_disable_irq(A1X_IRQ_TIMER0);
/* Attach the timer interrupt vector */
(void)irq_attach(A1X_IRQ_TIMER0, (xcpt_t)up_timerisr);
/* Enable interrupts from the TIMER 0 port */
regval = getreg32(A1X_TMR_IRQ_EN);
regval |= TMR_IRQ_TMR0;
putreg32(regval, A1X_TMR_IRQ_EN);
/* And enable the timer interrupt */
up_enable_irq(A1X_IRQ_TIMER0);
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,001
|
Q: socket.io, io is not defined (JS Error) I have just started with socket.io, its giving JS Error on client page
io is not defined
How to fix this ?
A: Alternatively you can use the Socket.io CDN:
<script src="https://cdn.socket.io/socket.io-1.0.0.js"></script>
A: put <script src="http://yournodeserver/socket.io/socket.io.js"></script> into your code
A: I have a socket app where my server (not a CDN) is serving up the socket.io.js script. So, while Emmerman is right in saying that you need to include the script tag in your client HTML code, the asset won't be loaded if your back-end is down. One option is for you to write a client-side JS script that checks for io before you try and use socket.io. If it's not present (undefined/null) then you can either conditionally show something else like, "server down" or in my case, I'm going to set a timer that keeps checking periodically until the server is restored.
[UPDATE 2] Ended up having to include the script tag, check for existence of io object and doing a window.location.reload() after 10 seconds (using setTimeout) (which eventually will hopefully find that the script loaded and io exists, after which I can connect to the socket server.)
[UPDATE] I'm loading the script with an ajax call rather than using an html script tag. Then with the timer I'm checking periodically if the script will load – eventually it will when the server is restored/rebooted. jQuery ref to load JS scripts dynamically: http://api.jquery.com/jQuery.getScript/
A: I faced the same problem when using express. Even putting the server:port inside the script would not work.After the server started i would make socket listen to that port, that was mistake i guess.Changing it to below works fine
var app = express();
app.set('port', process.env.PORT || 3000);
var server = http.createServer(app);
var io = require('socket.io').listen(server);
server.listen(app.get('port'));
On Client side I just include the script
<script src="/socket.io/socket.io.js"></script>
A: <script src="http://cdnjs.cloudflare.com/ajax/libs/socket.io/0.9.16/socket.io.min.js"></script>
this is the latest version of socket.io to be included.
A: Your make sure your using "server" instead "app".
const express = require('express');
const app = express();
const http = require('http');
const server = http.createServer(app);
const {
Server
} = require('socket.io');
const io = new Server(server);
app.get('/', (req, res) => {
res.sendFile(__dirname + '/index.html')
});
io.on('connection', (socket) => {
console.log('a user connected');
});
app.listen(3000, () => {
console.log('Listining on 3000');
}); // wrong
server.listen(3000, () => {
console.log('Listining on 3000');
}); // correct
A: Wrap your client code on a '$(document).ready()' for jQuery or another library similar function. This way you'll be sure your code runs after the library beeing loaded.
A: I had to do this. (For the client.)
function setup() {
var socket;
socket = io.connect('http://localhost:3000');
}
/*
This solved my error.
*/
A: If your server listening on port 3000 for example and your index.html connect on localhost:3000 then it will fail to include the local libraries on that port.
To solve this problem, you should have already installed a normal server for example xampp and include the full url for your library to be served from xampp into your page
for example : //localhost/my-project/node_modules/socket....
or just include it from cdn
or use socket = io.connect('http://localhost:3000'); and load your page from normal server for example xampp
A: Change order of scripts files, first library socket.io next yours scripts files
<script src="/socket.io/socket.io.js"></script>
<script src="./js/your-script.js"></script>
A: No need to edit anything just go to your index.js and make sure that you put :
server.listen(PORT,()=>{
/**Code**/
})
A: http://socket.io/download/ - the official page for latest cdn.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 3,465
|
package xenapi
import (
"fmt"
"github.com/amfranz/go-xmlrpc-client"
"reflect"
"strconv"
"time"
)
var _ = fmt.Errorf
var _ = xmlrpc.NewClient
var _ = reflect.TypeOf
var _ = strconv.Atoi
var _ = time.UTC
type VmssFrequency string
const (
// Hourly snapshots
VmssFrequencyHourly VmssFrequency = "hourly"
// Daily snapshots
VmssFrequencyDaily VmssFrequency = "daily"
// Weekly snapshots
VmssFrequencyWeekly VmssFrequency = "weekly"
)
type VmssType string
const (
// The snapshot is a disk snapshot
VmssTypeSnapshot VmssType = "snapshot"
// The snapshot is a checkpoint
VmssTypeCheckpoint VmssType = "checkpoint"
// The snapshot is a VSS
VmssTypeSnapshotWithQuiesce VmssType = "snapshot_with_quiesce"
)
type VMSSRecord struct {
// Unique identifier/object reference
UUID string
// a human-readable name
NameLabel string
// a notes field containing human-readable description
NameDescription string
// enable or disable this snapshot schedule
Enabled bool
// type of the snapshot schedule
Type VmssType
// maximum number of snapshots that should be stored at any time
RetainedSnapshots int
// frequency of taking snapshot from snapshot schedule
Frequency VmssFrequency
// schedule of the snapshot containing 'hour', 'min', 'days'. Date/time-related information is in Local Timezone
Schedule map[string]string
// time of the last snapshot
LastRunTime time.Time
// all VMs attached to this snapshot schedule
VMs []VMRef
}
type VMSSRef string
// VM Snapshot Schedule
type VMSSClass struct {
client *Client
}
// GetAllRecords Return a map of VMSS references to VMSS records for all VMSSs known to the system.
func (_class VMSSClass) GetAllRecords(sessionID SessionRef) (_retval map[VMSSRef]VMSSRecord, _err error) {
_method := "VMSS.get_all_records"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRefToVMSSRecordMapToGo(_method + " -> ", _result.Value)
return
}
// GetAll Return a list of all the VMSSs known to the system.
func (_class VMSSClass) GetAll(sessionID SessionRef) (_retval []VMSSRef, _err error) {
_method := "VMSS.get_all"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRefSetToGo(_method + " -> ", _result.Value)
return
}
// SetType
func (_class VMSSClass) SetType(sessionID SessionRef, self VMSSRef, value VmssType) (_err error) {
_method := "VMSS.set_type"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertEnumVmssTypeToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SetLastRunTime
func (_class VMSSClass) SetLastRunTime(sessionID SessionRef, self VMSSRef, value time.Time) (_err error) {
_method := "VMSS.set_last_run_time"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertTimeToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// RemoveFromSchedule
func (_class VMSSClass) RemoveFromSchedule(sessionID SessionRef, self VMSSRef, key string) (_err error) {
_method := "VMSS.remove_from_schedule"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_keyArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "key"), key)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _keyArg)
return
}
// AddToSchedule
func (_class VMSSClass) AddToSchedule(sessionID SessionRef, self VMSSRef, key string, value string) (_err error) {
_method := "VMSS.add_to_schedule"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_keyArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "key"), key)
if _err != nil {
return
}
_valueArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _keyArg, _valueArg)
return
}
// SetSchedule
func (_class VMSSClass) SetSchedule(sessionID SessionRef, self VMSSRef, value map[string]string) (_err error) {
_method := "VMSS.set_schedule"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertStringToStringMapToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SetFrequency Set the value of the frequency field
func (_class VMSSClass) SetFrequency(sessionID SessionRef, self VMSSRef, value VmssFrequency) (_err error) {
_method := "VMSS.set_frequency"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertEnumVmssFrequencyToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SetRetainedSnapshots
func (_class VMSSClass) SetRetainedSnapshots(sessionID SessionRef, self VMSSRef, value int) (_err error) {
_method := "VMSS.set_retained_snapshots"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertIntToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SnapshotNow This call executes the snapshot schedule immediately
func (_class VMSSClass) SnapshotNow(sessionID SessionRef, vmss VMSSRef) (_retval string, _err error) {
_method := "VMSS.snapshot_now"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_vmssArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "vmss"), vmss)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _vmssArg)
if _err != nil {
return
}
_retval, _err = convertStringToGo(_method + " -> ", _result.Value)
return
}
// SetEnabled Set the enabled field of the given VMSS.
func (_class VMSSClass) SetEnabled(sessionID SessionRef, self VMSSRef, value bool) (_err error) {
_method := "VMSS.set_enabled"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertBoolToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SetNameDescription Set the name/description field of the given VMSS.
func (_class VMSSClass) SetNameDescription(sessionID SessionRef, self VMSSRef, value string) (_err error) {
_method := "VMSS.set_name_description"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// SetNameLabel Set the name/label field of the given VMSS.
func (_class VMSSClass) SetNameLabel(sessionID SessionRef, self VMSSRef, value string) (_err error) {
_method := "VMSS.set_name_label"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_valueArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "value"), value)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg, _valueArg)
return
}
// GetVMs Get the VMs field of the given VMSS.
func (_class VMSSClass) GetVMs(sessionID SessionRef, self VMSSRef) (_retval []VMRef, _err error) {
_method := "VMSS.get_VMs"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertVMRefSetToGo(_method + " -> ", _result.Value)
return
}
// GetLastRunTime Get the last_run_time field of the given VMSS.
func (_class VMSSClass) GetLastRunTime(sessionID SessionRef, self VMSSRef) (_retval time.Time, _err error) {
_method := "VMSS.get_last_run_time"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertTimeToGo(_method + " -> ", _result.Value)
return
}
// GetSchedule Get the schedule field of the given VMSS.
func (_class VMSSClass) GetSchedule(sessionID SessionRef, self VMSSRef) (_retval map[string]string, _err error) {
_method := "VMSS.get_schedule"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertStringToStringMapToGo(_method + " -> ", _result.Value)
return
}
// GetFrequency Get the frequency field of the given VMSS.
func (_class VMSSClass) GetFrequency(sessionID SessionRef, self VMSSRef) (_retval VmssFrequency, _err error) {
_method := "VMSS.get_frequency"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertEnumVmssFrequencyToGo(_method + " -> ", _result.Value)
return
}
// GetRetainedSnapshots Get the retained_snapshots field of the given VMSS.
func (_class VMSSClass) GetRetainedSnapshots(sessionID SessionRef, self VMSSRef) (_retval int, _err error) {
_method := "VMSS.get_retained_snapshots"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertIntToGo(_method + " -> ", _result.Value)
return
}
// GetType Get the type field of the given VMSS.
func (_class VMSSClass) GetType(sessionID SessionRef, self VMSSRef) (_retval VmssType, _err error) {
_method := "VMSS.get_type"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertEnumVmssTypeToGo(_method + " -> ", _result.Value)
return
}
// GetEnabled Get the enabled field of the given VMSS.
func (_class VMSSClass) GetEnabled(sessionID SessionRef, self VMSSRef) (_retval bool, _err error) {
_method := "VMSS.get_enabled"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertBoolToGo(_method + " -> ", _result.Value)
return
}
// GetNameDescription Get the name/description field of the given VMSS.
func (_class VMSSClass) GetNameDescription(sessionID SessionRef, self VMSSRef) (_retval string, _err error) {
_method := "VMSS.get_name_description"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertStringToGo(_method + " -> ", _result.Value)
return
}
// GetNameLabel Get the name/label field of the given VMSS.
func (_class VMSSClass) GetNameLabel(sessionID SessionRef, self VMSSRef) (_retval string, _err error) {
_method := "VMSS.get_name_label"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertStringToGo(_method + " -> ", _result.Value)
return
}
// GetUUID Get the uuid field of the given VMSS.
func (_class VMSSClass) GetUUID(sessionID SessionRef, self VMSSRef) (_retval string, _err error) {
_method := "VMSS.get_uuid"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertStringToGo(_method + " -> ", _result.Value)
return
}
// GetByNameLabel Get all the VMSS instances with the given label.
func (_class VMSSClass) GetByNameLabel(sessionID SessionRef, label string) (_retval []VMSSRef, _err error) {
_method := "VMSS.get_by_name_label"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_labelArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "label"), label)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _labelArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRefSetToGo(_method + " -> ", _result.Value)
return
}
// Destroy Destroy the specified VMSS instance.
func (_class VMSSClass) Destroy(sessionID SessionRef, self VMSSRef) (_err error) {
_method := "VMSS.destroy"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_, _err = _class.client.APICall(_method, _sessionIDArg, _selfArg)
return
}
// Create Create a new VMSS instance, and return its handle. The constructor args are: name_label, name_description, enabled, type*, retained_snapshots, frequency*, schedule (* = non-optional).
func (_class VMSSClass) Create(sessionID SessionRef, args VMSSRecord) (_retval VMSSRef, _err error) {
_method := "VMSS.create"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_argsArg, _err := convertVMSSRecordToXen(fmt.Sprintf("%s(%s)", _method, "args"), args)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _argsArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRefToGo(_method + " -> ", _result.Value)
return
}
// GetByUUID Get a reference to the VMSS instance with the specified UUID.
func (_class VMSSClass) GetByUUID(sessionID SessionRef, uuid string) (_retval VMSSRef, _err error) {
_method := "VMSS.get_by_uuid"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_uuidArg, _err := convertStringToXen(fmt.Sprintf("%s(%s)", _method, "uuid"), uuid)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _uuidArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRefToGo(_method + " -> ", _result.Value)
return
}
// GetRecord Get a record containing the current state of the given VMSS.
func (_class VMSSClass) GetRecord(sessionID SessionRef, self VMSSRef) (_retval VMSSRecord, _err error) {
_method := "VMSS.get_record"
_sessionIDArg, _err := convertSessionRefToXen(fmt.Sprintf("%s(%s)", _method, "session_id"), sessionID)
if _err != nil {
return
}
_selfArg, _err := convertVMSSRefToXen(fmt.Sprintf("%s(%s)", _method, "self"), self)
if _err != nil {
return
}
_result, _err := _class.client.APICall(_method, _sessionIDArg, _selfArg)
if _err != nil {
return
}
_retval, _err = convertVMSSRecordToGo(_method + " -> ", _result.Value)
return
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 7,846
|
Q: PHP clear up Array preg_grep I have currently a problem with arrays in php,
i requested all <li> tags from an html site with:
$dochtml = new DOMDocument();
$dochtml -> loadHTMLFile("my.html");
/*get all li's*/
$lis = $dochtml -> getElementsByTagName('li');
My Html Body:
<li>gfvgvefg</li>
<li>ferfergegeg</li>
<li id="vdfvdfvf"></li>
<li id="id_1"> My Value 1</li>
<li id="id_2"> My Value 2</li>
<li id="id_3"> My Value 3</li>
<li id="id_4"> My Value 4</li>
<li></li>
<li id="efgvefgvfg">gfegvefgbv</li>
then i echo all id's and values from my <li> tags:
foreach($lis as $key) {
$id = $key -> getAttribute('id');
$value = $key -> nodeValue;
echo $id. ' - '.$value.'<br>';
}
Output:
0 - gfvgvefg
0 - ferfergegeg
vdfvdfvf - 0
id_1 - My Value 1
id_2 - My Value 2
id_3 - My Value 3
id_4 - My Value 4
0 - 0
efgvefgvfg - gfegvefgbv
You see some id's are 0, and values too.
What i want is this Output (print_r):
Array ( [id_1] => My Value 1
[id_2] => My Value 2
[id_3] => My Value 3
[id_4] => My Value 4)
So only the id's with id_* and the correct Value, in an array.
I try this:
foreach($lis as $key) {
// gets, and outputs the ID and content of each DIV
$id = $key -> getAttribute('id');
$value = $key -> nodeValue;
//echo $id. ' - '.$value.'<br>';
$array = array($value => $id);
$matches = preg_grep('/^id_(\w+)/i', $array);
print_r($matches);
}
but i get this output (print_r):
Array ( ) Array ( ) Array ( ) Array ( ) Array ( ) Array ( ) Array ( ) Array ( ) Array ( )
no id no value, nothing.
I know i can use array_keys by preg_grep, but then i get only the id's and the empty arrays…
Here is the code in an online compile, for better understanding.
Code Online Compiled
Help me please.
A: The same using XPath:
$dom =new DOMDocument;
$dom->loadHTMLFile("my.html");
$xp = new DOMXPath($dom);
$nodeList = $xp->query('//li[starts-with(@id, "id_")]');
foreach ($nodeList as $node) {
$result[$node->getAttribute('id')] = $node->nodeValue;
}
print_r($result);
A: Do everything in a single foreach:
$arrLis = array();
foreach($lis as $key) {
$id = $key -> getAttribute('id');
// check if `id` starts with `id_` string
if (strpos($id, 'id_') === 0) {
$value = $key -> nodeValue;
echo $id. ' - '.$value.'<br>';
$arrLis[$id] = $value;
}
}
print_r($arrLis);
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,789
|
{"url":"https:\/\/aviation.stackexchange.com\/tags\/density-altitude\/hot","text":"# Tag Info\n\n20\n\nLow density altitudes mean your engine generates more power, and your wings generate more lift. Everything is better with low density altitudes, right? The potential danger in low density altitudes is that your engine generates more power than it can handle. Even without exceeding the maximum rated RPM and manifold pressure, you can exceed the rated power ...\n\n16\n\nWith more flaps, you will lift off at lower speed, but will have worse climb performance due to the increased drag. On short field you are limited by the field length, so you need to lift off early and therefore need to use more flaps. But in the mountains you are limited by the climb performance. Therefore you need to use less flaps. Of course, this is ...\n\n13\n\nThe optimum altitude depends on your aircraft, the engines and the weather. But most important is to decide what you are trying to achieve. If you would like to optimize your fuel consumption per distance travelled, the altitude you will fly at will be higher than if you try to optimize your fuel consumption per time unit. Fuel per distance travelled is ...\n\n12\n\nThere are procedures with temperature restrictions, related to altitude constraints. An example is Innsbruck: The text in the red box says: Procedure N\/A below AD temp -7\u00b0 For effect of temperature on altimeter: How will the altimeter read in air colder than ISA?\n\n11\n\nAn oversimplification for sure: Turbochargers essentially compress the air going into the engine (cylinders) to maintain (near) sea level pressure internally, so the external density altitude doesn't affect the performance of the engine. The \"liability\" at low density altutudes is that it is possible to over-boost the engine, potentially damaging it. See ...\n\n11\n\nThe atmosphere approximates an ideal gas, and as such you can relate pressure and density through the ideal gas equation. The form we use in meteorology uses mass density and is given by: $$p={\\rho}RT$$ where $P$ is pressure in units of Pa, $\\rho$ is density in units of kg m-3, $R$ is the gas constant for dry air (287 J kg-1 K-1) and $T$ is temperature in ...\n\n11\n\n(Source) The basics Height: absolute height above ground, measured using a radar altimeter (like the helicopter on the right) QFE: the barometric setting an airport broadcasts that makes an altimeter on ground read zero. This gives height only around the airport if the area is relatively flat. Also known as Above Field Elevation QNH: the barometric setting ...\n\n10\n\nThe air in the lungs is in pressure equilibrium with the cabin, but not temperature equilibrium. It's always at body temperature. So the only variable that can affect how much oxygen is available to the body is the pressure of the outside air - thus pressure altitude, not density altitude, is physiologically relevant. That said, if the density altitude is ...\n\n10\n\nGenerally, climbing as high as you can would be close to the optimum strategy. In the end, the altitude is given by the optimum lift coefficient in cruise, which can be approximated as $$c_L = \\sqrt{\\frac{1-n_v}{n_v+3}\\cdot \\pi \\cdot AR \\cdot \\epsilon \\cdot c_{D0}}$$ Now it will be important to get the thrust over speed dependency right, which is expressed ...\n\n10\n\nYes, through the Reynolds number The Reynolds Number is a function of airspeed and is used to predict the transition from laminar to turbuluent flow: $${Re} = \\frac{\\rho v L}{\\mu} = \\frac{v L}{\\nu}$$ The critical angle of attack is affected by the separation of the boundary layer, which, in turn, depends on the laminar or turbulent qualities of said ...\n\n8\n\nThe caution is probably the default prefix for a nonstandard DA notice, not anything particular to the fact that it's negative. You're correct that lower DA will improve performance, and an FAAsafety.gov document on density altitude (PDF) doesn't discuss low DA at all.\n\n8\n\nQNH is the atmospheric pressure adjusted to sea level. So, a reported QNH of 29.92 means the pressure is 29.92 inHg at sea level, regardless of the reporting station's elevation. If you're on ground at 3,000' MSL, you can rotate the knob to display 0' altitude, the pressure reading then on the altimeter window will be QFE (Field Elevation). (Source) Real ...\n\n8\n\nIf the altimeter says the station pressure is 29.92\" sea level, but I am at 3000', then the airplane still thinks it's at sea level when it comes to performance. Is that wrong? Although the station pressure may be reported as 29.92\", that does not mean that a barometer there will show that as the current pressure. The reported value is the actual pressure, ...\n\n7\n\nAssuming our lungs are like our engines They are not. Engine (at least spark-ignited reciprocating one) uses most oxygen in each charge and its RPM is limited, so it can't be run faster to compensate for the lower air density. However, our bodies only use small fraction of oxygen from each breath and we normally breathe much slower than we can if need be. ...\n\n6\n\nI have analyzed this information before, and this is what the chart looks like with one line for each weight category if you convert the temperature and pressure altitudes to density altitudes and then chart that vs. the takeoff distance: You can see that I have drawn a sort of best-fit line for this information. I did not add humidity as a factor because ...\n\n5\n\nWe all know the saying \"From High to Low, Look out below\" Some were taught that this only applied to pressure differences and thus were instructed to get a new altimeter setting every 100NM or so. This is a good recommendation. Temperature differences from standard also causes errors with the altimeter. The same saying above holds true with temperature ...\n\n5\n\nCheck your temperature conversion: 25F is -3.9C... Should be (25-32)*(5\/9).\n\n5\n\nYou should look in the POH for your specific aircraft (because it may have had modifications to it which would make it different from a \"standard\" aircraft) and see what it says. The 172P POH that I found online contains no maximum airport elevation for takeoff, and the Type Certificate Data Sheet (TCDS) doesn't either. Therefore, you can only go by the ...\n\n5\n\nThe International Standard Atmosphere (ISA) gives \u201caverage\u201d values of air pressure and density by altitude. It starts by defining standard pressure 1013.25 hPa (29.92 inHg) and standard temperature 15\u00b0C at sea level, and derives the values for higher altitude from physical laws. A \u201cdensity altitude\u201d is altitude in ISA at which air has the density you are ...\n\n5\n\nThere is a variety of reasons for a maximum altitude: Most common is insufficient engine performance to climb any higher. Then your question never arises. The aircraft simply cannot go higher. For more powerful aircraft the partial pressure of the oxygen supply in an unpressurized cabin would drop below the value to keep the pilot(s) alert. For GA aircraft ...\n\n5\n\nYou may be able to get a rough estimate of if your engine is producing sufficient thrust utilizing brake pressure. But, how will you determine how much thrust is being produced? Your brakes are not calibrated so carefully as to adequately determine that. You would only be guessing. You might get a better indication of engine performance by utilizing the ...\n\n4\n\nI don't know of a maximum airfield elevation, though I suppose the airplane's service ceiling would qualify as an answer to this question. In regards to the second question, that is largely going to depend on the airfield elevation, runway length, climb gradients needed for obstacle clearance, and ambient atmospheric conditions. These questions would have ...\n\n4\n\nPressure, density and temperature are related (approximately) through the ideal gas equation. In the general form it is $$PV=nRT$$ Where $P$ is pressure, $V$ is volume, $n$ is amount, $T$ is temperature and $R$ is ideal gas constant. If you have an enclosed container filled with air, volume ($V$) and amount ($n$) are the same, so pressure increases ...\n\n4\n\nA big thing to remember is that $Density=\\frac{Mass}{Volume}$. It is unrelated to pressure, and pressure is unrelated to density. Pressure generally increases with temperature only in a gas with a constant volume. This is because you are adding more energy to the system, causing the molecules to become more exited. To put simply, they bounce around harder ...\n\n4\n\nThe mistake is subtracting the ISA temperature at sea level from the ISA+18\u00b0F temperature at altitude in step two. If your atmosphere is heated up, it needs to be hotter at sea level as well, or your lapse rate in the first step and the denominator is wrong. First calculate density for ISA conditions. This gives 0.7708 kg\/m\u00b3. Then used the ideal gas law to ...\n\n4\n\nThe formula can be derived equaling the weight ($W$) of an airplane with the maximum lift ($L$) generated by its wings. I'm not aware of any special name for it. $$W = L \\\\ m\\cdot g= \\frac{\\rho_{min}}{2}\\cdot V^2\\cdot S\\cdot c_{L,max}$$ Substituting the aircraft's speed with an expression containing the Mach number ($a$ is the local speed of sound), V=...\n\n4\n\nAdd 10% to your computed takeoff distance if it is too humid. Humidity. Humidity is not generally considered a major factor in density altitude computations because the effect of humidity is related to engine power rather than aerodynamic efficiency. At high ambient temperatures, the atmosphere can retain a high water vapor content. For example, at 96\u00b0F, ...\n\n4\n\nYou would have to have a brake pressure indicator that shows the pressure being applied to the calipers, where you could observe the minimum pressure required to hold the plane when WOT, and some charts or calculations that showed the static thrust being produced with a given amount of brake pressure. But then the amount of pressure varies with a lot of ...\n\n3\n\nAnother factor in the decision to fly low or high: turbulence. How important is it to you and your passengers to have a smooth ride? Higher is generally less turbulent but check those Airmets and Pireps!\n\n3\n\nTurbochargers increase available engine performance by increasing the amount of fuel\/air mixture that can be burned in the cylinders. This is accomplished by compressing the air before fuel is mixed in. The performance of a reciprocating aircraft engine is directly dependent on how much fuel can be burned and converted into heat energy, which in turn is ...\n\nOnly top voted, non community-wiki answers of a minimum length are eligible","date":"2021-02-27 12:50:22","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 1, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.6211210489273071, \"perplexity\": 965.1026744546788}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2021-10\/segments\/1614178358956.39\/warc\/CC-MAIN-20210227114444-20210227144444-00589.warc.gz\"}"}
| null | null |
\section{Introduction}
Motivated by the potential social impact and fueled by the recent advances in both hardware (sensor technologies such as lidar) and software (artificial intelligence techniques such as deep learning), massive efforts from both industry and academia have been invested to autonomous driving during the last decade. Start from the DARPA urban challenges~\cite{urmson2008autonomous,montemerlo2008junior}, a number of autonomous vehicle system demonstrations have been performed. Automotive industry giants such as Benz~\cite{ziegler2014making}, and IT giants such as Google~\cite{bansal2018chauffeurnet} are competing to develop the first commercial fully autonomous vehicle.
A typical autonomous driving system is organized as a two parts architecture~\cite{badue2019self,yurtsever2019survey}: a perception system, and a decision making system, as shown in Fig.\ref{Fig:architecture}. The perception system is composed of multiple subsystems including detection~\cite{yang2018pixor}, tracking~\cite{luo2014multiple}, localization and mapping~\cite{bresson2017simultaneous}. These systems, together with offline collected maps, transform the raw sensor inputs (e.g, camera RGB images and lidar point clouds) to useful information such as surrounding vehicles' poses, ego vehicle's pose, and the local semantic map centered around the ego vehicle. On the other hand, the decision making system is divided into subsystems including routing~\cite{bauer2010combining}, behavior prediction~\cite{tang2019adaptive}, decision \& planning~\cite{chen2018continuous,chen2019autonomous}, and control~\cite{paden2016survey} that work together to generate a control command (e.g, steering angle, throttle and braking) to drive the autonomous car.
\begin{figure}
\centering
\includegraphics[width = .48\textwidth]{architecture.png}
\caption{\label{Fig:architecture}Typical architecture of autonomous driving systems, which is composed of the perception part and decision making part with multiple subsystems.}
\end{figure}
Although this highly modularized architecture works well in a few driving tasks, it starts to touch its performance limitations because (1) too much human heuristics can lead to inappropriate perception results and driving behaviors; and (2) Too many complicated subsystems are making the whole system expensive to scale and maintain. Alternatively, end-to-end architectures might avoid those limitations, as the driving models can be learned and continuously optimized from data, without much hand-engineered involvement. However, although efforts have been made to build such autonomous driving architectures~\cite{chen2020interpretable,chen2019model,chen2019deep}, an end-to-end system with decision making in the loop can only be demonstrated in simulations, real world applications are still infeasible due to safety issues. On the other hand, the perception system alone is suitable to be designed in an end-to-end form, since it does not require online driving exploration.
In this paper, we propose an end-to-end approach for the autonomous driving perception system defined in Fig.\ref{Fig:architecture}. This method enables us to solve the detection, tracking, localization and mapping problems altogether, by learning a sequential latent representation model. The learned model is able to simultaneously provide accurate estimation of surrounding vehicle poses, ego vehicle global pose, and local semantic roadmap. With this end-to-end approach, we only need minimum human engineering efforts to obtain a fully functional perception system, and no maps are needed online. Furthermore, the fusion of these subsystems helps improve the performance of surrounding vehicle pose estimation.
The remainder of this paper is organized as follows. Section~\ref{sec:related} summarizes existing works for autonomous driving perception. Section~\ref{sec:pre} analyzes the subsystems of a typical perception system to help us better understand their purposes and principles. Details of our proposed method is introduced in section~\ref{sec:proposed}. Section~\ref{sec:experiments} shows the experiments and results, while section~\ref{sec:conclusion} concludes the paper.
\section{Related Work}\label{sec:related}
It is crucial to perceive the environment and extract useful information. This mainly includes vehicle detection, tracking, localization and mapping.
\subsection{Vehicle Detection}
Vehicle detection refers to estimating the position, heading and size of surrounding vehicles. There are three main subclasses: (1) Image-based vehicle detection generates bounding boxes on front-view camera image~\cite{redmon2016you,ren2015faster}; (2) Semantic segmentation assigns each pixel of the image with a class label. Pixels belonging to same objects have the same label~\cite{long2015fully,he2017mask}; and (3) 3D object detection obtains the 3D poses (or bird-eye poses) of surrounding vehicles, which is usually achieved with the help of lidar point cloud~\cite{chen2017multi,yan2018second}.
\subsection{Vehicle Tracking}
Direct prediction from the vehicle detection system is often insufficient, more accurate vehicle state needs to be estimated given historical detection. Typical vehicle tracking methods have two phases: (1) data association to connect objects between frames~\cite{hwang2016fast,nguyen2011stereo}; and (2) filtering methods such as Kalman Filters and Particle Filters to smooth the vehicle dynamics~\cite{ess2010object,petrovskaya2009model}.
\subsection{Localization and Mapping}
Localization is the task of estimating ego vehicle pose relative to a reference frame in a map, which can be either a raw point cloud map or an annotated semantic map, depending on the algorithm we choose. There are two main approaches: (1) Simultaneous localization and mapping (SLAM) makes map online and localize the ego vehicle in the map at the same time~\cite{bresson2017simultaneous}; and (2) A priori map-based localization that estimate the ego vehicle pose by finding the best match to a detailed a priori map~\cite{levinson2007map}.
\section{Preliminary}\label{sec:pre}
In this section, we will analyze the typical subsystems in an autonomous driving perception system, including detection, tracking, localization and mapping. They are reformulated into graphical models. This helps us better understand their purposes and relationships to fuse them into a single end-to-end framework.
\subsection{Typical Detection and Tracking Systems}
The goal of the detection and tracking subsystems is to accurately estimate the states of surrounding vehicles. This includes their positions and heading angles relative to the ego vehicle, as well as their length and width. Current detection and tracking systems are divided into two subsystems. First, the detection subsystem predicts the vehicles' poses based on single frame sensor inputs. Then, the tracking subsystem smooths the results from the detection system based on its historical outputs.
These processes can be interpreted as a graphical model, as shown in Fig~\ref{Fig:detection}. The red block represents the detection subsystem, which contains a detection model $p\left(\hat{d}_t|x_t\right)$ that maps the sensor inputs $x_t$ to an estimation of the surrounding vehicles' poses $\hat{d}_t$. This part is usually performed by fitting a deep neural network model using supervised learning techniques. The green block represents the tracking subsystem, which estimates the true vehicles' poses $d_t$ given historical outputs of the detection system $\hat{d}_{1:t}$ and sometimes the historical ego vehicle actions $a_{1:t}$. This estimation problem can be formulated as a filter problem $p\left(d_t|\hat{d}_{1:t},a_{1:t}\right)$ which is usually solved by Kalman filters with hard-coded transition model $p\left(d_{t+1}|d_t,a_t\right)$ and observation model $p\left(\hat{d}_t|d_t\right)$.
\begin{figure}
\centering
\includegraphics[width = .42\textwidth]{detection.png}
\caption{\label{Fig:detection}Graphical model of typical detection and tracking systems. The red block represents the detection system. The green block represents the tracking system.}
\end{figure}
\subsection{Typical Localization and Mapping Systems}
The localization and mapping system needs to accurately estimate the pose of the ego vehicle in the coordinate of a global map. Then a local semantic map indicating road geometry, topology and traffics will be obtained, which are used for downstream planning and control tasks. The estimated global ego vehicle pose is also used for routing.
These processes can be interpreted as a graphical model, as shown in Fig~\ref{Fig:slam}. The global ego vehicle pose $l_t$ is estimated based on historical sensor inputs and actions $p\left(l_t|x_{1:t}, a_{1:t}\right)$. The global map $m$ can be edited offline or constructed online, depending on the methodology we use. After estimating the ego vehicle pose, a local semantic map $S^{loc}_t$ is obtained by locating the ego vehicle on the global map, which requires semantic annotations. Note that the global map $m$ needs to be stored online.
\begin{figure}
\centering
\includegraphics[width = .43\textwidth]{slam.png}
\caption{\label{Fig:slam}Graphical model of typical localization and mapping system.}
\end{figure}
\section{Proposed Method}\label{sec:proposed}
In this paper, we propose a novel end-to-end autonomous driving perception system for simultaneous detection, tracking, localization and mapping. All the functionalities are fused in a single framework. A sequential latent representation learning process is performed to learn this end-to-end perception model.
\subsection{End-to-end Autonomous Driving Perception}
In general, the purpose of detection and tracking is to estimate surrounding vehicles' poses $d_t$, while the purpose of localization and mapping is to obtain the local semantic map $S^{loc}_t$ and the global ego vehicle pose $l_t$. All these estimations are conditioned on historical sensor inputs $x_{1:t}$ and actions $a_{1:t}$. Therefore, the tasks of perception can be simplified by estimating the following conditional probability:
\begin{equation}\label{Eq:task}
p\left(d_t,S_t^{loc},l_t|x_{1:t},a_{1:t}\right)
\end{equation}
As stated in section~\ref{sec:pre}, typical methods estimate~\eqref{Eq:task} by dividing it into multiple separate tasks, and tackle them one-by-one with lots of human engineering efforts. Different from typical methods, we propose to solve the problem of estimating~\eqref{Eq:task} jointly. This is possible by assuming that there is a latent space summarizing all useful historical information. Then with this latent space, we can extract the information we need, such as surrounding vehicles' poses, road geometry, and ego vehicle pose. Inspired by works that learn latent representations with time sequence reasoning~\cite{krishnan2015deep,lee2019stochastic,chen2020interpretable}, we propose to formulate the end-to-end perception system as a single graphical model, as shown in Fig.~\ref{Fig:e2e}. A more detailed architecture of our model in a single frame is shown in Fig.~\ref{Fig:arch}
\begin{figure}
\centering
\includegraphics[width = .45\textwidth]{e2e.png}
\caption{\label{Fig:e2e}Graphical model of end-to-end autonomous driving perception.}
\end{figure}
\begin{figure*}
\centering
\includegraphics[width = .95\textwidth]{arch.png}
\caption{\label{Fig:arch}Architecture of our proposed end-to-end perception model at a single frame.}
\end{figure*}
$z_t$ is the latent variable we introduce, it represents a summary of historical information and is evolved with the latent dynamics $p\left(z_{t+1}|z_t,a_t\right)$. All relevant information about the environment are decoded from this latent, including the sensor inputs $p\left(x_t|z_t\right)$, surrounding vehicles' poses $p\left(d_t|z_t\right)$, local semantic roadmap $p\left(S_t^{loc}|z_t\right)$ and global ego vehicle pose $p\left(l_t|z_t\right)$. If we are able to estimate the distribution of the latent state given historical sensor inputs and actions $q\left(z_t|x_{1:t},a_{1:t}\right)$, then \eqref{Eq:task} can be obtained by integrating out the latent state:
\begin{equation}
\begin{aligned}
&p\left(d_t,S_t^{loc},l_t|x_{1:t},a_{1:t} \right)\\
&\qquad\qquad=\int p\left(d_t,S_t^{loc},l_t|z_t \right) q\left(z_t|x_{1:t},a_{1:t}\right) d z_t
\end{aligned}
\end{equation}
Generally we do not need to calculate the exact integration, but rather its expectation, which can be approximated by sampling:
\begin{equation}
\hat{z}_t \sim q\left(z_t|x_{1:t},a_{1:t}\right) \quad \hat{d}_t,\hat{S}_t^{loc},\hat{l}_t \sim p\left(d_t,S_t^{loc},l_t|\hat{z}_t \right)
\end{equation}
The next subsection will introduce how we can learn a model to estimate $q\left(z_t|x_{1:t},a_{1:t}\right)$ and $p\left(d_t,S_t^{loc},l_t|z_t \right)$ with sequential latent representation learning.
\subsection{Sequential Latent Representation Learning}
To learn appropriate models shown in Fig.\ref{Fig:e2e}, we need to fit them with collected dataset. For convenience, we first denote a trajectory to be composed of sensor inputs, detection outputs, local semantic roadmaps, ego vehicle poses and actions:
\begin{equation}
\vec{x}=x_{1:t},\,\vec{d}=d_{1:t},\,\vec{S^{loc}}=S^{loc}_{1:t},\,\vec{l}=l_{1:t},\,\vec{a}=a_{1:t}
\end{equation}
The dataset is then composed of this kind of trajectories collected while driving ${\cal D} = \left \{ \left(\vec{x}^i, \vec{d}^i, \left(\vec{S^{loc}}\right)^i, \vec{l}^i, \vec{a}^i \right ) \right \}_{i=1}^N$. The model can be fitted by maximizing the log likelihood of the data:
\begin{equation}
\begin{aligned}
&\textrm{log}\, \prod_{i=1}^N p\left (\vec{x}^i, \vec{d}^i, \left(\vec{S^{loc}}\right)^i, \vec{l}^i \left|\right. \vec{a}^i\right ) \\
&\qquad \qquad \qquad =\sum_{i=1}^N \textrm{log}\,p\left (\vec{x}^i, \vec{d}^i, \left(\vec{S^{loc}}\right)^i, \vec{l}^i \left|\right. \vec{a}^i\right )
\end{aligned}
\end{equation}
which can be maximized using stochastic gradient descent (SGD), which optimizes parametric functions with gradient descent. The gradient is estimated by sampling a batch of data points. To apply SGD to our problem, $p\left (\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l} \left|\right. \vec{a}\right )$ needs to be composed of parametric functions, thus auto-differentiation tools such as TensorFlow can be used to evaluate their gradients. Variational inference~\cite{kingma2013auto} can be applied to compute this log likelihood. First, introduce the latent variables $\vec{z}=z_{1:t}$:
\begin{equation}\label{Eq:latent}
\textrm{log}\,p\left (\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l} \left|\right.\vec{a}\right ) =\textrm{log}\,\int p\left(\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l}, \vec{z} \left|\right.\vec{a}\right) d \vec{z}
\end{equation}
Then introduce a variational distribution $q\left(\vec{z}|\vec{x},\vec{a}\right)$ into \eqref{Eq:latent}:
\begin{equation}\label{Eq:vardist}
\begin{aligned}
&\textrm{log}\,p\left (\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l} \left|\right.\vec{a}\right ) \\
&\qquad \qquad =\textrm{log}\,\int p\left(\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l}, \vec{z} \left|\right.\vec{a}\right) \frac{q\left(\vec{z}|\vec{x},\vec{a}\right)}{q\left(\vec{z}|\vec{x},\vec{a}\right)} d \vec{z}
\end{aligned}
\end{equation}
Now eliminate the integration in \eqref{Eq:vardist} by introducing expectation, and then apply Jensen's inequality:
\begin{equation}\label{Eq:ELBO}
\begin{aligned}
\textrm{log}\,p\left(\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l}|\vec{a} \right)
&=\textrm{log}\underset{q\left(\vec{z}|\vec{x},\vec{a}\right)}{\mathbb{E}}\left[\frac{p\left(\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l}, \vec{z}|\vec{a} \right)}{q\left(\vec{z}|\vec{x},\vec{a}\right)}\right]\\
&\geq \underset{q\left(\vec{z}|\vec{x},\vec{a}\right)}{\mathbb{E}}\left[\textrm{log}\,p\left(\vec{x}, \vec{d}, \vec{S^{loc}}, \vec{l}, \vec{z}|\vec{a} \right)\right.\\
&\qquad \qquad \qquad \qquad \left.-\textrm{log}\,q\left(\vec{z}|\vec{x},\vec{a}\right)\right]\\
&=\text{ELBO}
\end{aligned}
\end{equation}
where ELBO stands for evidence lower bound. The original log likelihood can be maximized by maximizing this ELBO. Now derive $p\left(\vec{x},\vec{d},\vec{S^{loc}},\vec{l},\vec{z}|\vec{a} \right)$ by probability factorization according to the PGM in Fig.\ref{Fig:e2e}:
\begin{equation}\label{Eq:factor}
\begin{aligned}
&p\left(\vec{x},\vec{d},\vec{S^{loc}},\vec{l},\vec{z}|\vec{a} \right)\\
&\qquad \qquad=p\left(\vec{x},\vec{d},\vec{S^{loc}},\vec{l}|\vec{z},\vec{a} \right)p\left(\vec{z}|\vec{a}\right)\\
&\qquad \qquad=p\left(\vec{x}|\vec{z}\right)p\left(\vec{d}|\vec{z}\right)p\left(\vec{S^{loc}}|\vec{z}\right)p\left(\vec{l}|\vec{z}\right)p\left(\vec{z}|\vec{a}\right)
\end{aligned}
\end{equation}
And substitute \eqref{Eq:factor} into \eqref{Eq:ELBO}, we have:
\begin{equation}\label{Eq:modellearning}
\begin{aligned}
&\textrm{ELBO}=\underset{q\left(\vec{z}|\vec{x},\vec{a}\right)}{\mathbb{E}}[\textrm{log}\,p\left(\vec{x}|\vec{z}\right)+\textrm{log}\,p\left(\vec{d}|\vec{z}\right)+\textrm{log}\,p\left(\vec{S^{loc}}|\vec{z}\right)\\
&\qquad \qquad \qquad +\textrm{log}\,p\left(\vec{l}|\vec{z}\right)+\textrm{log}\,p\left(\vec{z}|\vec{a}\right) -\textrm{log}\,q\left(\vec{z}|\vec{x},\vec{a}\right)]
\end{aligned}
\end{equation}
Now derive the components in \eqref{Eq:modellearning} by unfolding them with time. Considering the conditional dependence of PGM in Fig.\ref{Fig:e2e}. The decoding models can be unfolded as:
\begin{equation}
\begin{aligned}
&\textrm{log}\,p\left(\vec{x}|\vec{z}\right)=\textrm{log}\prod_{\tau=1}^{t}p\left(x_\tau|z_\tau\right)=\sum_{\tau=1}^{t}\textrm{log}\,p\left(x_\tau|z_\tau\right)\\
&\textrm{log}\,p\left(\vec{d}|\vec{z}\right)=\textrm{log}\prod_{\tau=1}^{t}p\left(d_\tau|z_\tau\right)=\sum_{\tau=1}^{t}\textrm{log}\,p\left(d_\tau|z_\tau\right)\\
&\textrm{log}\,p\left(\vec{S^{loc}}|\vec{z}\right)=\textrm{log}\prod_{\tau=1}^{t}p\left(S^{loc}_\tau|z_\tau\right)=\sum_{\tau=1}^{t}\textrm{log}\,p\left(S^{loc}_\tau|z_\tau\right)\\
&\textrm{log}\,p\left(\vec{l}|\vec{z}\right)=\textrm{log}\prod_{\tau=1}^{t}p\left(l_\tau|z_\tau\right)=\sum_{\tau=1}^{t}\textrm{log}\,p\left(l_\tau|z_\tau\right)\\
\end{aligned}
\end{equation}
The prior model can be unfolded using the latent state transition function:
\begin{equation}
\begin{aligned}
\textrm{log}\,p\left(\vec{z}|\vec{a}\right)
&=\textrm{log}\left[p\left(z_1\right)\prod_{\tau=1}^{t-1} p\left(z_{\tau+1}|z_\tau,a_\tau\right)\right]\\
&=\textrm{log}\,p\left(z_1\right)+\sum_{\tau=1}^{t-1} \textrm{log}\,p\left(z_{\tau+1}|z_\tau,a_\tau\right)
\end{aligned}
\end{equation}
The latent state inference model can be unfolded as:
\begin{equation}
\begin{aligned}
\textrm{log}\,q\left(\vec{z}|\vec{x},\vec{a}\right)
&=\textrm{log}\left[q\left(z_1|\vec{x},\vec{a}\right)\prod_{\tau=1}^{t-1} q\left(z_{\tau+1}|z_\tau,\vec{x},\vec{a}\right)\right]\\
&\approx \textrm{log}\left[q\left(z_1|x_1\right)\prod_{\tau=1}^{t-1} q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)\right]\\
&=\textrm{log}\,q\left(z_1|x_1\right)+\sum_{\tau=1}^{t-1}\textrm{log}\, q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)
\end{aligned}
\end{equation}
Note here we approximate $q\left(\vec{z}|\vec{x},\vec{a}\right)$ and $q\left(z_{\tau+1}|z_\tau,\vec{x},\vec{a}\right)$ with $q\left(z_1|x_1\right)$ and $q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)$ for simplicity. To obtain the exact values, bi-directional recurrent neural networks should be used to obtain the posterior probabilities conditioned on the whole trajectory sequence $\left(\vec{x},\,\vec{a}\right)$~\cite{krishnan2015deep}.
Now we can unfold \eqref{Eq:modellearning} with time:
\begin{equation}\label{Eq:modelELBO}
\begin{aligned}
&\textrm{ELBO} \approx \underset{q\left(\vec{z}|\vec{x},\vec{a}\right)}{\mathbb{E}}\left[\sum_{\tau=1}^t\textrm{log}\,p\left(x_\tau|z_\tau\right)+\sum_{\tau=1}^t\textrm{log}\,p\left(d_\tau|z_\tau\right)\right.\\
&\qquad+ \sum_{\tau=1}^t\textrm{log}\,p\left(S^{loc}_\tau|z_\tau\right) +\sum_{\tau=1}^t\textrm{log}\,p\left(l_\tau|z_\tau\right)\\
&\qquad-\text{D}_{\text{KL}}\left(q\left(z_1|x_1\right)||p\left(z_1\right)\right)\\
&\qquad \left.-\sum_{\tau=1}^t\text{D}_{\text{KL}}\left(q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)||p\left(z_{\tau+1}|z_\tau,a_\tau\right)\right)\right]
\end{aligned}
\end{equation}
and the ELBO is now decomposed to several simple networks, which will be illustrated in section~\ref{sec:architecture}.
\subsection{Input Representations}
\subsubsection{\bf{Sensor Input $x$}}
We use two sensors to provide the observations, camera and lidar. For camera, the sensor input is a front-view RGB image, which can be represented by a tensor of $[0, 255]^{128\times 128\times 3}$, as shown in Fig.\ref{Fig:input}(a). For lidar, we project the point clouds to the ground plane and render them into a 2D lidar image. The lidar image is represented by a tensor of $[0, 255]^{128\times 128\times 3}$, with each pixel rendered in red or green depending on whether there are lidar points at or above ground level existing in the corresponding pixel cell, as shown in Fig.\ref{Fig:input}(b).
We use camera and lidar together because they are both important sensor sources and provide complementary information. Lidar point clouds provides accurate spatial information of other road participants and obstacles in 360 degrees of view. While the front-view camera is good at providing information of the road conditions.
\begin{figure}
\centering
\includegraphics[width = .45\textwidth]{input.png}
\caption{\label{Fig:input}Input representations. (a): Front camera RGB image; (b) lidar birdeye image; (c) local semantic roadmap. All images are with shape $128\times 128$.}
\end{figure}
\subsubsection{\bf{Detection Mask $d$}}
The detection mask is composed of two branches: a classification branch outputing a 1-channel classification feature map and a regression branch outputing a 6-channel regression feature map. Both feature maps is in bird-eye view and their representation follows that in~\cite{yang2018pixor}. Briefly speaking, the classification feature map indicates the probability of each pixel belonging to a surrounding vehicle. While the regression feature map is composed of geometry information of the corresponding surrounding vehicle, e.g, x-y positions, heading, width and length of each detected vehicle (See Section 3.1.1 in \cite{yang2018pixor} for details). With these two feature maps, we can use Non-Maximum Suppression (NMS) to get the final detection results. Thus $d$ is composed of two tensors of $\mathbb{R}^{128\times 128\times 1}$ and $\mathbb{R}^{128\times 128\times 6}$ respectively.
\subsubsection{\bf{Local Semantic Roadmap $S^{loc}$}}
The local semantic roadmap is a map centered around the ego vehicle which includes semantic information about the road geometry, road topology and traffic rules. For example, the lane markings, drivable areas, and stop signs. It is an RGB image of $[0, 255]^{128\times 128\times 3}$, as shown in Fig.\ref{Fig:input}(c).
\subsubsection{\bf{Global Ego Vehicle State $l$}}
The global ego vehicle pose includes the information of ego vehicle's x-y position (in meter) and heading angle (in rad) in the global map's coordinate. It is a vector of $\mathbb{R}^{3}$.
\subsection{Network Architectures}\label{sec:architecture}
In this section, we will illustrate the detailed architectures of the networks in \eqref{Eq:modelELBO}.
\subsubsection{\bf{Sequential Latent Model}}
The sequential latent model includes the latent dynamics network $p\left(z_{\tau+1}|z_\tau,a_\tau\right)$, the filtering model network $q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)$, $q\left(z_1|x_1\right)$, and the sensor inputs reconstruction network $p\left(x_\tau|z_\tau\right)$. Here we follow the two-layer hierarchical latent space structure as in \cite{lee2019stochastic}, such that $z_\tau=[z_\tau^1, z_\tau^2] \in \mathbb{R}^{288}$ where $z_\tau^1 \in \mathbb{R}^{32}$ and $z_\tau^2 \in \mathbb{R}^{256}$. $p\left(z_{\tau+1}|z_\tau,a_\tau\right)$ consists of two fully connected layers with hidden units number 256, followed by a Gaussian output layer. $q\left(z_{\tau+1}|z_\tau,x_{\tau+1},a_\tau\right)$ and $q\left(z_1|x_1\right)$ both consist of 5 convolutional layers ((32, 5, 2), (64, 3, 2), (128, 3, 2), (256, 3, 2), (256, 3, 2) and (256, 4, 1), with each tuple means (filters, kernel size, strides), as shown in Fig.\ref{Fig:networks}(a)) to first encode the sensor inputs $x_t$ into features of size 256. Then two fully connected layers with hidden units number 256 are followed, with a Gaussian output layer. $p\left(x_\tau|z_\tau\right)$ both consist of 5 deconvolutional layers ((256, 4, 1), (256, 3, 2), (128, 3, 2), (64, 3, 2), (32, 3, 2), and (3, 5, 2), with each tuple means (filters, kernel size, strides), as shown in Fig.\ref{Fig:networks}(b)) with a fixed standard deviation of 0.1.
\begin{figure}
\centering
\includegraphics[width = .48\textwidth]{networks.png}
\caption{\label{Fig:networks}Network architectures. (a): Image encoder for camera and lidar images; (b) Image decoder for camera, lidar and roadmap images; (c) Decoder for detection masks.}
\end{figure}
\subsubsection{\bf{Detection and Tracking Network}}
The detection and tracking network is the network generating the detection mask $p\left(d_\tau|z_\tau\right)$. We deployed network architecture that is similar to the network architecture in \cite{yang2018pixor}. The mask, including a one-channel tensor and a six-channel, is decoded from the latent by 4 deconvolutional layers ((256, 4, 1), (128, 3, 2), (64, 3, 2), (32, 3, 2), with each tuple means (filters, kernel size, strides)), and then a deconvolutional layer of (1, 5, 2) and (6, 5, 2) respectively, as shown in Fig.\ref{Fig:networks}(c).
\subsubsection{\bf{Localization and Mapping Network}}
The localization and mapping network includes the local semantic roadmap decoder $p\left(S^{loc}_\tau|z_\tau\right)$ and the global ego vehicle state decoder $p\left(l_\tau|z_\tau\right)$. $p\left(S^{loc}_\tau|z_\tau\right)$ has the same architecture with $p\left(x_\tau|z_\tau\right)$, as shown in Fig.\ref{Fig:networks}(b). $p\left(l_\tau|z_\tau\right)$ is a two-layer fully connected neural network with hidden units number 256.
\section{Experiments}\label{sec:experiments}
\subsection{Simulation Setup and Data Collection}\label{sec:simulation}
We train and evaluate our proposed method on CARLA~\cite{dosovitskiy2017carla}. CARLA simulator is a high-definition open-source simulation platform designed for autonomous driving research. It simulates not only the driving environment and vehicle dynamics, but also the raw sensor data inputs such as camera RGB images and lidar point clouds. Fig.\ref{Fig:carla} (a) shows a sample view of the driving simulation environment we use.
\begin{figure}
\centering
\begin{subfigure}[b]{0.22\textwidth}
\centering
\includegraphics[height = \linewidth]{carla_view.png}
\caption{Sample view of CARLA simulator}
\end{subfigure}
~~
\begin{subfigure}[b]{0.22\textwidth}
\centering
\includegraphics[height = \linewidth]{map.png}
\caption{Map layout of the simulated city}
\end{subfigure}
\caption{\label{Fig:carla}Simulation environment}
\end{figure}
To collect the data for training, we navigate the ego vehicle in virtual town of CARLA. Fig.\ref{Fig:carla} (b) shows the map layout of the town. It includes various urban scenarios such as intersections and roundabouts. The range of the map is $400m \times 400m$, with about $6km$ total length of roads. 100 vehicles are running autonomously in the virtual town to simulate a crowded urban environment. Both the ego vehicle and the surrounding vehicles will randomly choose a direction at intersections, then follow the route, while slowing down for front vehicles and stopping when the front traffic light becomes red. We run the ego vehicle for 50k environment steps and store the observations of each step into the training dataset.
\subsection{Training Details}
The model is trained with a batch size of 32 and learning rate 0.0001. The length of sequential model used for training is $t = 10$. The total iteration of training is 100k. We train three variants of our methods:
\subsubsection{\bf{Inputs and roadmap}}
Both the raw sensor inputs and the local semantic roadmap are reconstructed when training. So the model is enforced to capture the features of both sensor inputs and the local semantic roadmap.
\subsubsection{\bf{No inputs reconstruction}}
No raw sensor inputs are reconstructed. This is reasonable since we do not necessary need to output the reconstructed raw sensor inputs.
\subsubsection{\bf{No roadmap reconstruction}}
No local semantic roadmaps are reconstructed. Then the model is not enforced to capture the features of the road geometry and traffic rules.
\subsection{Evaluation Results}\label{sec:eval}
To evaluate our method, instead of evaluating on a collected test dataset, we directly put the ego vehicle in a random start point in the virtual town with surrounding vehicles, as described in \ref{sec:simulation}. During the navigation, the ego vehicle is collecting sensor inputs, encoding it to latent states, and decoding to perception outputs. We let the vehicle runs for 15k environment steps and evaluate the performance during this period of navigation.
Fig.\ref{Fig:example} shows examples of the perception output. The first row represents the original camera and lidar inputs, as well as the ground truth local semantic roadmap and surrounding vehicle bounding boxes. The second row shows the perception output given only the historical camera and lidar inputs. We can see that the model is able to generate accurate semantic roadmap and surrounding vehicle bounding boxes, even for the occluded area.
\begin{figure*}
\centering
\includegraphics[width = 1.0\textwidth]{example.png}
\caption{\label{Fig:example}Example results. The first row shows sensor inputs and ground truth, second row shows reconstructed outputs. Left to right: camera image, lidar image, local semantic roadmap, surrounding vehicle bounding boxes. Note that only camera and lidar inputs are given for the reconstruction, the ground truth semantic roadmap and bounding boxes are displayed here only for comparison.}
\end{figure*}
We also evaluate the statistic performance of the system according to typical evaluation metrics. For surrounding vehicle bounding box prediction, we plot the Precision-Recall Curve (PRC) and then compute the Average Precision (AP) as Area Under Precision-Recall Curve (AUC)~\cite{everingham2010pascal}. The PRC and APs are computed under Intersection-Over-Union (IoU) of 0.1, 0.3, 0.5, and 0.7. Fig.\ref{Fig:PRC} shows the PRC of the methods. The APs are summarized in Table.\ref{Tab:AP}. we can see the variant that does not reconstruct the semantic roadmap has significantly worse performance than the other two variants. This shows that fusing the information of map might improve the performance of detection.
\begin{figure}
\centering
\includegraphics[width = .5\textwidth]{PRC.png}
\caption{\label{Fig:PRC}Precision-Recall Curves for surrounding vehicles bounding boxes prediction.}
\end{figure}
\begin{table}
\centering
\caption{Average precision for surrounding vehicles bounding boxes prediction.\label{Tab:AP}}
\resizebox{0.48\textwidth}{!}{
\begin{tabular}{c|cccc}
\hline
& $AP_{0.1}$ & $AP_{0.3}$ & $AP_{0.5}$ & $AP_{0.7}$ \\ \hline
Inputs and roadmap & \bf{79.4}\% & 72.0\% & 56.5\% & 16.8\% \\
No inputs reconstruction & 78.4 & \bf{74.4\%} & \bf{61.0\%} & \bf{22.1\%} \\
No roadmap reconstruction & 57.4\% & 45.3\% & 7.8\% & 0.3 \% \\\hline
\end{tabular}}
\end{table}
For ego vehicle global state, we calculate its average prediction error. There are two values we care about, the location error (in meter) and heading error (in rad). Table.\ref{Tab:error} shows the evaluation results. We can see that we get an average global location error of 7.1 meters, and heading error of 0.17 rad. This is obtained purely from the raw camera and lidar sensor inputs, with no GPS or stored global map used.
\begin{table}
\centering
\caption{Average error of ego vehicle global pose estimation.\label{Tab:error}}
\resizebox{0.43\textwidth}{!}{
\begin{tabular}{c|cc}
\hline
& Location (m) & Heading (rad) \\ \hline
Inputs and roadmap & 11.4 & 0.33 \\
No inputs reconstruction & 8.6 & \bf{0.17} \\
No roadmap reconstruction & \bf{7.1} & 0.42 \\\hline
\end{tabular}}
\end{table}
\section{Conclusion and Future Works}\label{sec:conclusion}
In this paper, we proposed and implemented an end-to-end autonomous driving perception system based on sequential latent representation learning. The system was able to replace the functionalities of the typical detection, tracking, localization and mapping subsytems with minimum human engineering efforts and without online stored maps. The method was evaluated in a realistic autonomous driving simulator, taking camera RGB image and lidar point cloud as sensor inputs. Evaluation results showed the learned model can obtain accurate surrounding vehicles' poses, local semantic roadmaps, and global ego vehicle pose based purely on historical raw sensor inputs.
Although the learned model performs reasonably well, it has a large space for improvement. The neural network architectures used in this paper are only the very basic ones, such as shallow convolutional layers. The size of input images is only $128 \times 128$, making it very hard to detect objects that are small or far away. In the future, we will deploy more advanced network architectures, and train the model on images with higher definition. We will also test this system with a downstream autonomous driving decision making system.
\bibliographystyle{ieee}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 6,111
|
Q: Indexing into containers: the mathematical underpinnings When you want to pull an element out of a data structure, you have to give its index. But the meaning of index depends on the data structure itself.
class Indexed f where
type Ix f
(!) :: f a -> Ix f -> Maybe a -- indices can be out of bounds
For example...
Elements in a list have numeric positions.
data Nat = Z | S Nat
instance Indexed [] where
type Ix [] = Nat
[] ! _ = Nothing
(x:_) ! Z = Just x
(_:xs) ! (S n) = xs ! n
Elements in a binary tree are identified by a sequence of directions.
data Tree a = Leaf | Node (Tree a) a (Tree a)
data TreeIx = Stop | GoL TreeIx | GoR TreeIx -- equivalently [Bool]
instance Indexed Tree where
type Ix Tree = TreeIx
Leaf ! _ = Nothing
Node l x r ! Stop = Just x
Node l x r ! GoL i = l ! i
Node l x r ! GoR j = r ! j
Looking for something in a rose tree entails stepping down the levels one at a time by selecting a tree from the forest at each level.
data Rose a = Rose a [Rose a] -- I don't even like rosé
data RoseIx = Top | Down Nat RoseIx -- equivalently [Nat]
instance Indexed Rose where
type Ix Rose = RoseIx
Rose x ts ! Top = Just x
Rose x ts ! Down i j = ts ! i >>= (! j)
It seems that the index of a product type is a sum (telling you which arm of the product to look at), the index of an element is the unit type, and the index of a nested type is a product (telling you where to look in the nested type). Sums seem to be the only one which aren't somehow linked to the derivative. The index of a sum is also a sum - it tells you which part of the sum the user is hoping to find, and if that expectation is violated you're left with a handful of Nothing.
In fact I had some success implementing ! generically for functors defined as the fixed point of a polynomial bifunctor. I won't go into detail, but Fix f can be made an instance of Indexed when f is an instance of Indexed2...
class Indexed2 f where
type IxA f
type IxB f
ixA :: f a b -> IxA f -> Maybe a
ixB :: f a b -> IxB f -> Maybe b
... and it turns out you can define an instance of Indexed2 for each of the bifunctor building blocks.
But what's really going on? What is the underlying relationship between a functor and its index? How does it relate to the functor's derivative? Does one need to understand the theory of containers (which I don't, really) to answer this question?
A: It seems like the index into the type is an index into the set of constructors, following by an index into the product representing that constructor. This can be implemented quite naturally with e.g. generics-sop.
First you need a datatype to represent possible indices into a single element of the product. This could be an index pointing to an element of type a,
or an index pointing to something of type g b - which requires an index pointing into g and an index pointing to an element of type a in b. This is encoded with the following type:
import Generics.SOP
data ArgIx f x x' where
Here :: ArgIx f x x
There :: (Generic (g x')) => Ix g -> ArgIx f x x' -> ArgIx f x (g x')
newtype Ix f = ...
The index itself is just a sum (implemented by NS for n-ary sum) of sums over the generic representation of the type (choice of constructor, choice of constructor element):
newtype Ix f = MkIx (forall x . NS (NS (ArgIx f x)) (Code (f x)))
You can write smart constructors for various indices:
listIx :: Natural -> Ix []
listIx 0 = MkIx $ S $ Z $ Z Here
listIx k = MkIx $ S $ Z $ S $ Z $ There (listIx (k-1)) Here
treeIx :: [Bool] -> Ix Tree
treeIx [] = MkIx $ S $ Z $ S $ Z Here
treeIx (b:bs) =
case b of
True -> MkIx $ S $ Z $ Z $ There (treeIx bs) Here
False -> MkIx $ S $ Z $ S $ S $ Z $ There (treeIx bs) Here
roseIx :: [Natural] -> Ix Rose
roseIx [] = MkIx $ Z $ Z Here
roseIx (k:ks) = MkIx $ Z $ S $ Z $ There (listIx k) (There (roseIx ks) Here)
Note that e.g. in the list case, you cannot construct an (non-bottom) index pointing to the [] constructor - likewise for Tree and Empty, or constructors containing values whose type is not a or something containing some values of type a. The quantification in MkIx prevents the construction bad things like an index pointing to the first Int in data X x = X Int x where x is instantiated to Int.
The implementation of the index function is fairly straightforward, even if the types are scary:
(!) :: (Generic (f x)) => f x -> Ix f -> Maybe x
(!) arg (MkIx ix) = go (unSOP $ from arg) ix where
atIx :: a -> ArgIx f x a -> Maybe x
atIx a Here = Just a
atIx a (There ix0 ix1) = a ! ix0 >>= flip atIx ix1
go :: (All SListI xss) => NS (NP I) xss -> NS (NS (ArgIx f x)) xss -> Maybe x
go (Z a) (Z b) = hcollapse $ hzipWith (\(I x) -> K . atIx x) a b
go (S x) (S x') = go x x'
go Z{} S{} = Nothing
go S{} Z{} = Nothing
The go function compares the constructor pointed to by the index and the actual constructor used by the type. If the constructors don't match, the indexing returns Nothing. If they do, the actual indexing is done - which is trivial in the case that the index points exactly Here, and in the case of some substructure, both indexing operations must succeed one after the other, which is handled by >>=.
And a simple test:
>map (("hello" !) . listIx) [0..5]
[Just 'h',Just 'e',Just 'l',Just 'l',Just 'o',Nothing]
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 733
|
How to manage the human consequences of Syria's chaos
Order from Chaos
Elizabeth Ferris and Kemal Kirişci Tuesday, April 26, 2016
Half of Syria's population is now either an internally displaced person (IDP) or a refugee. The international community is still struggling to respond to this humanitarian catastrophe, now in its sixth year, with the resources of host governments, aid agencies, nongovernmental organizations, and a multitude of other actors stretched to the limit. In the absence of viable political solutions to the Syrian conflict, no end to the war is in sight—and funds to support an ever-increasing population of displaced persons are neither sufficient now nor likely to be sustainable in the future. Displaced Syrians—whom we should see not just as victims but as survivors—are taking matters into their own hands, leaving Syria in massive numbers and making treacherous journeys to seek safety elsewhere.
The Syrian tragedy is occurring against the backdrop of a global displacement crisis. According to the United Nations High Commissioner for Refugees, almost 60 million people have been forced from their homes by conflict, violence, and persecution. And a large proportion doesn't go home quickly. Finding solutions for long-term displacement has been on the global humanitarian agenda for years, but the international community is failing in this task, especially in terms of providing adequate support for refugees and IDPs to become self-reliant, resettling refugees, and planning for sustainable solutions.
All trends point to Syria becoming yet another protracted displacement crisis with profound implications for Jordan, Lebanon, and Turkey, the major host countries for Syrian refugees—as well as the EU and the broader international community. Can something be done about it?
Failed by our international structures?
We need to first recognize the implications of Syrian displacement for the international order. The ongoing conflict and the displacement it has caused demonstrate that international structures designed to prevent and resolve conflicts—including the United Nations and regional bodies—are not working, at least not in the Middle East.
Ultimately, the solution to the Syrian IDP and refugee crisis is a political one, since it requires an end to the violence and destruction in Syria. It was only in late-2015, when the Syria conflict began to affect the security and stability of Europe, that high-level efforts were intensified to address the root cause of displacement. So far, a highly fragile and partial truce on the ground, paired with efforts by U.N. Envoy Staffan de Mistura to keep the warring parties at the negotiation table in Geneva, have not in any way lessened the challenges of extending humanitarian assistance to displaced Syrians. The dire humanitarian situation continues to call for burden sharing and international solidarity.
The dire humanitarian situation continues to call for burden sharing and international solidarity.
To date, the burden of protecting and assisting refugees has largely fallen on the shoulders of major host countries—primarily Jordan, Lebanon, and Turkey. The governments of these countries are providing a public good for the international community, and it is indeed disappointing that appeals by host governments and the U.N. system for more assistance received such scant attention—until large numbers of Syrians began to arrive in Europe. Even now, the resources that have been marshaled aren't sufficient for the gargantuan task at hand, though Europe and the international community are realizing that they need a new formula for better burden sharing with host countries in Syria's neighborhood.
A how-to
It is against this background that we propose a New Global Approach for Syria in our new book (part of the forthcoming Brookings Marshall Paper series), "Consequences of Chaos." This approach would bring together the governments of refugee-hosting countries, the U.N. and other intergovernmental agencies, regional bodies, international nongovernmental organizations, local civil society actors, and donor governments to consider and adopt a new system of burden sharing. The focus would be:
Reaffirming the principle that protecting refugees is an international responsibility;
Supporting common legal and policy approaches to Syrian refugees in the region that includes access to livelihood opportunities;
Reaffirming resettlement as a core component of refugee protection and assistance and re-tooling elements of resettlement policy to meet the needs of the most vulnerable refugees, in particular;
Providing a forum for creative thinking on solutions for internally displaced people;
Establishing a new relationship between humanitarian and development actors;
Engaging development actors such as the World Bank more effectively; and
Laying the groundwork for longer-term reconstruction and recovery efforts in Syria.
We propose that developing this New Global Approach for Syria could be worked out through a consultative process with stakeholders over a six-to-twelve-month period. It would be jointly led by the U.N. secretary general and the president of the World Bank, and could culminate in a global meeting in early 2017. There is no shortage of creative ideas for strengthening all of these elements—from enhancing resettlement to strengthening coordination between humanitarian and development agencies. The main challenge, as always, comes in implementation, to be carried out by a range of actors, including by national governments and international organizations. If successfully implemented, the New Global Approach for Syria would offer a win-win outcome: foremost for Syrian refugees and IDPs, but also for major host countries as well as the EU, not to mention the broader international community. This new system of combining relief and humanitarian assistance with a developmental approach may form the skeleton of a template for managing the broader global refugee crisis, as well as help reform international humanitarian governance.
Turkey and the West
By Kemal Kirişci
The Consequences of Chaos
By Elizabeth G. Ferris and Kemal Kirişci
By Bruce Riedel
Some tentative steps have already been taken towards implementing such an approach. In early February, the "Supporting Syria and the Region" conference in London raised over $11 billion in pledges. In March, the UNHCR held a high-level meeting calling on governments around the world to substantially increase their own resettlement programs for Syrian refugees. A deal struck in March between the European Union and Turkey has received considerable criticism but might mark a turning point in terms of regional engagement with the issue. However, what is still lacking is a comprehensive and well-coordinated approach to addressing the Syrian displacement crisis that is in some sync with efforts to consolidate the truce in Syria, ensure humanitarian access to affected populations, and push for a political solution to the conflict.
Elizabeth Ferris
Former Brookings Expert
Research Professor, Institute of Study of International Migration - Georgetown University
Twitter Beth_Ferris
Kemal Kirişci
TÜSİAD Senior Fellow - Foreign Policy, Center on the United States and Europe
Director - The Turkey Project
Twitter kemalkirisci
A how-to guide for managing the end of the post-Cold War era. Read all the Order from Chaos content »
Humanitarian & Disaster Assistance
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,993
|
"""Classes supporting dynamically registering custom modules."""
__author__ = 'Pavel Simakov (psimakov@google.com)'
All_LOCALES_PERMISSION = 'can_pick_all_locales'
All_LOCALES_DESCRIPTION = 'Can pick all locales, including unavailable ones.'
SEE_DRAFTS_PERMISSION = 'can_see_draft_content'
SEE_DRAFTS_DESCRIPTION = 'Can see lessons and assessments with draft status.'
import collections
import roles
class Module(object):
"""A class that holds module information."""
def __init__(
self, name, desc, global_routes, namespaced_routes,
notify_module_enabled=None, notify_module_disabled=None):
self._name = name
self._desc = desc
self._global_routes = global_routes
self._namespaced_routes = namespaced_routes
self._notify_module_enabled = notify_module_enabled
self._notify_module_disabled = (
notify_module_disabled or Module.module_disabling_is_deprecated)
Registry.registered_modules[self._name] = self
def disable(self):
raise NotImplementedError('Disabling modules is not supported.')
@staticmethod
def module_disabling_is_deprecated():
raise NotImplementedError('Disabling modules is not supported.')
def enable(self):
Registry.enabled_module_names.add(self.name)
if self._notify_module_enabled:
self._notify_module_enabled()
@property
def enabled(self):
return self.name in Registry.enabled_module_names
@property
def name(self):
return self._name
@property
def desc(self):
return self._desc
@property
def global_routes(self):
if self.name in Registry.enabled_module_names:
return self._global_routes
else:
return []
@property
def namespaced_routes(self):
if self.name in Registry.enabled_module_names:
return self._namespaced_routes
else:
return []
class Registry(object):
"""A registry that holds all custom modules."""
registered_modules = collections.OrderedDict()
enabled_module_names = set()
@classmethod
def get_all_routes(cls):
global_routes = []
namespaced_routes = []
for registered_module in cls.registered_modules.values():
if registered_module.enabled:
# Only populate the routing table with enabled modules.
global_routes += registered_module.global_routes
namespaced_routes += registered_module.namespaced_routes
return global_routes, namespaced_routes
core_module = None
def register_core_module(global_handlers, namespaced_handlers):
"""Creates module containing core functionality.
This is not really a module, in the sense that it's not optional.
However, the items present are much more conveniently implemented
by using the Module logic. Thus, rather than putting something
in coursebuilder/modules/core, we have this here, so as to indicate
that it's not really a module in the broader sense.
"""
global core_module # pylint: disable=global-statement
def permissions_callback(unused_app_context):
return [
roles.Permission(All_LOCALES_PERMISSION, All_LOCALES_DESCRIPTION),
roles.Permission(SEE_DRAFTS_PERMISSION, SEE_DRAFTS_DESCRIPTION)
]
def notify_module_enabled():
roles.Roles.register_permissions(core_module, permissions_callback)
core_module = Module(
'Core REST services', 'A module to host core REST services',
global_handlers, namespaced_handlers,
notify_module_enabled=notify_module_enabled)
core_module.enable()
def can_pick_all_locales(app_context):
return roles.Roles.is_user_allowed(
app_context, core_module, All_LOCALES_PERMISSION)
def can_see_drafts(app_context):
return roles.Roles.is_user_allowed(
app_context, core_module, SEE_DRAFTS_PERMISSION)
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 7,388
|
The Federal Government is getting set to implement the recommendations of the Justice Obiora Nwazota Judicial Commission which probed the operations of the liquidated Nigeria Airways Limited (NAL), is a very positive and welcome development.
We hope to see action on this as soon as possible because the looting of our national flag carrier was one of the most outstanding cases of mindless corruption in our history.
unded The airline which was founded in 1958 once boasted of 32 airworthy aircraft which serviced domestic and international routes, competing with some of the best airlines in Africa and Asia in the late 1970s. But by the early 1980s, gross mismanagement, overstaffing and corruption had set in.
After the KLM technical partners hired to run the airline as a profitable business left in the early 1980s, it was a steady downhill roll, especially after the military took over power in 1983.
An airline that lifted 2.1 million passengers in 1985 could not carry more than 10,000 in 2003 with its sole surviving aircraft.
The Federal Government liquidated it in 2004. In 2002, the Justice Nwazota panel was set up to probe the operations of the company.
After a year of work, the panel turned in a four-volume report which revealed mindboggling corruption and looting to the tune of over N60 billion.
After the Federal Government approved the White Paper on this report and directed the Federal Ministry of Finance and the Police to recover the stolen funds and prosecute the indicted culprits, nothing more was heard about it.
Meanwhile, thousands of staff of the Airways who were thrown out of their jobs could not get paid their salaries and pensions. They languished in hunger and many died.
It was only recently that the Federal Government announced the approval of N22.5 billion to offset their estimated 45 billion wages and pension arrears.
One of the reasons that the President Muhammadu Buhari government could not proceed with the floating of a new national flag carrier was the need to settle outstanding issues from the liquidated Nigeria Airways.
Certainly, the recovery of whatever remains of its stolen funds and property and the punishment of the offenders are top among the unfinished business. We call on the government to go ahead with its plan to bring the looters of the rested Nigeria Airways to book.
It will help restore confidence in the government's anti-graft war, especially as this case has nothing to do with politics. Most of those indicted by the Justice Nwazota panel are still around active in the aviation sector.
Their prosecution will strengthen the confidence of future investors in our aviation sector such that when the second national flag carrier is eventually unveiled, stakeholders will be aware that the law is watching.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,794
|
"The Bath": This story is about communication, or rather, how bad we are at it, and yet how important it seems to be. Track the various attempts to communicate and their outcomes. What sort of failures can you find, and what are the consequences? How might things have been different if everyone was able to express themselves more effectively? The baker example is the obvious one. I'm interested in more subtle exchanges.
"Tell the Women We're Going": This is an unusually violent Carver story; at least, the violence more directly concerns the main characters. What else is comparatively odd about this story, in terms of point of view, characterization, use of time, and so on?
"After the Denim" is subtle compared to the other stories mentioned above. And yet, one of the characters is clearly facing serious medical problems. Is there any relief for the bad luck of these rather sympathetic characters? Seems to me there is, but is it enough?
"So Much Water So Close to Home": Here is a story that might be compared to "Tell the Women…" The main male character here is not a murderer, of course, but he's cold, and it seems the point of view character is suddenly seeing that men in general are threatening, are all potential monsters. How can you explain the sudden turn at the very end of the story? Why does the wife react to her husband's advance like this?
"The Third Thing that Killed My Father Off": If men can be evil, this story seems to suggest, so can women.My question about this story is about narration. The object of the story, Dummy, doesn't get to tell his own story. The father, who also is part of the story, doesn't either. Dummy's wife certainly has no say in things, or the narrator's mother. So why is the kid, Jack, telling this story, which really doesn't have a whole lot to do with him? What affect might the story have on Jack? What does it say about him that he apparently needs to tell this story?
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,983
|
Maternal Health Mapping: Join the Community, Put Your Organization on the Map!
The Maternal Health Task Force currently has three interactive (and fast growing!) maternal health maps; maternal health knowledge hubs, MPH programs that offer a concentration/focus on maternal health, and maternal health organizations. As part of my role at the MHTF, I am working to spread the word about these maps and further populate them so that they can serve as useful resources for maternal health professionals and students around the world. See below for a recent blog post on the MHTF Blog with info on how to put your organization or school of public health or medicine on the maps!
"Using geographic maps provides a global view of where maternal health activity is occurring and helps the community to understand where gaps exist. So far, we have three interactive maps on our site containing information that we hope will be useful in your work or studies; maternal health knowledge hubs, MPH programs that offer a concentration/focus on maternal health, and maternal health organizations.
At the upcoming Women Deliver conference in Washington D.C. (June 7-9), we will be engaging conference participants in our mapping activities and encouraging them to put their organization or school on the map! If you are attending Women Deliver, please visit our table in the Registration Hall to map your organization or school and to see a demo of how our mapping system works.
Putting your organization on our maternal health map will help to build a growing interconnected community of maternal health organizations. It will link the important work of your organization or school with the work of maternal health organizations and schools around the globe.
Also, be sure to check out Maternova's maternal health mapping activities! Maternova is working to map clinics and hospitals with maternal health services around the globe.
Click here to visit the Maternova maternal health map–and learn more about the initiative.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 9,184
|
If you speak to anyone who has attained a high level of mastery in their chosen field of study they will often make reference to their inspiration, influences and teachers that guided them on their own journey of exploration. During my process and still today, I hold a deep respect for the sacred tradition of this intimate relationship of mentorship.
Below are some definitions of terms used from different cultures to describe that relationship.
Sensei: (Japanese, that is literally translated as person born before another). In general usage, it is used, with proper form, after a person's name, and means "teacher", A title to refer to or address figures of authority. To show respect to someone who has achieved a certain level of mastery in an art form or some other skill: martial arts, novelists, musicians, artists, priests, or a roshi.
Swami: (Hindi, meaning master, lord, prince), used as a term of respectful address, is an ascetic or yogi who has been initiated into the religious monastic order founded by some religious teacher.
Guru: (Sanskrit, "GU" means shadows and "RU" refers to one who has the power to disperse them. therefore the Guru is seen as the one who "dispels the darkness of ignorance.") teacher, Master, educator of wisdom. In Hinduism, the guru is considered a respected person with saintly qualities who enlightens the mind of his or her disciple. In the guru/disciple relationship subtle and advanced knowledge is conveyed and received through the student's respect, commitment, devotion and obedience. The student eventually masters the knowledge that the guru embodies.
More recently in the world of pop-culture we even see the brave "super hero" rigorously train the trusty sidekick in the ways of fighting crime and of course who can forget one of the most famous movie renditions of the wise Jedi master who helps to awaken and focus the force within the young Padawan. This age-old tradition is also getting a modern make-over and revival in our secular world into its current incarnation of the "Life Coach." This label often refers to an old tradition of one who offers tools, insight and guides the student to master their own life circumstances inside and out to become a conscious, active participant and creator in their own personal life narrative.
There are many terms used in multiple traditions to describe not only this idea of a mentor or coach. The essence of this intimate teacher/student relationship imparts mystical knowledge and wisdom, illuminates the unconscious and often radiates a powerful "spiritual" transmission or invitation of exploring and knowing your Self at a deeper level of Awareness that is truly beyond the ability of the rational mind to comprehend.
For the last 20 years of my life I have always had one or more teachers, guides or mentors who play this role for me. Without their honest, objective perspective and their deep love and devotion I know that I would not have realized the version of who I am today. This is why I advocate, endorse, and personally practice this alchemical relationship (as teacher & student) and I firmly believe in the power that a true coach, mentor or a guide offers to your spiritual growth. At the highest level, there is a humility, knowingness and mature recognition when a "student" or seeker realizes that internal hunger for something more—deeper refinement, training, and growth through an objective example, role model or living example of what they respect, admire and hope to emulate.
The teacher is the one who is commissioned for the sole purpose to help empower you to realize, celebrate and express the highest and most evolved version of you, one who is devoted to aid in the path of Self-discovery at some level and assists in the refinement of those foundational character traits of accountability, integrity, balance, humility and respect within you.
Everyone needs an objective and alternative perspective that is truly working to serve your highest potential. This relationship aids to clarify, reflect, and inspire you to put into action the necessary steps to transform your inner and outer relationship. It assists you to "let go" of the old, stagnant and contractive patterns that no longer serve your life and it also inspires you to expand beyond your perceived limits to realize and live more of your unlimited potential. If you can envision something MORE, OTHER, & BEYOND then you are not only ready, but you have already begun the process to realize that relationship within and without.
The teacher's presence "affects" you. You will resonate to your teacher or mentor and you will feel a sense of kinship with their manner, teaching method and their very presence. The mentor will feel strangely familiar and their presence in your life will ignite something within you that is ready to come to the surface. There will be an undeniable connection that may not make rational sense, yet some inexplicable sense of "je ne sais quoi" makes your feel strangely attuned. The teacher may offer an invitation but the teacher rarely chooses the student. The student chooses the teacher.
The teacher serves your Self (big "S") not your self (small "s"). A truly conscious teacher is devoted and honest with you and will lovingly challenge and inspire you to be "uncomfortable" and push you to the far end of your comfort zone where all lasting transformation occurs. They will not be fooled by your excuses, seduced by your patterns and they will recognize and sometimes confront the mercurial and sly ego-matrix that is often your worst enemy. Their job is to hold you accountable to the truth and empower your process. The teacher can only do this for you, because they have learned to see this in their own consciousness. They cannot truly help you to transform anything in you that they have not already mastered within.
A true guide empowers you through Expansion. Expansion inspires expansion… An effective teacher will inspire, uplift, honor and encourage you through expansion as opposed to contraction. They will point out unrefined parts of you that may make you uncomfortable, but their focus will be on helping you heal, transcend and integrate those unrefined aspects. We often experience powerful expansion as discomfort as we push outside of our familiar and stuck spaces, but that is different than contraction. People often confuse the two. When you realize the difference between contraction and discomfort then real growth will prevail. The teacher can guide you through recognizing this process.
The honorable teacher will "walk his talk." This is another way of saying that your mentor has personal integrity. They reflect on the outside what is on the inside. Your guide will not only reflect and teach their version of the truth, but they are also be a living example of their own teachings and process. They teach from their personal experience and offer you the tools, guidance and perspective that they have practiced themselves and have achieved a level of mastery with in their own journey.
A great teacher is continually devoted to their own evolvement. A great teacher was (and is) a devoted student. Choose a teacher who is always hungry for more growth… They cherish and consciously invite refinement with humility and grace. They are the epitome of the perpetual student of life, always looking to refine themselves and deepen their experience of Self. Be cautious of anyone who espouses that they have "arrived" or are "done." Humility is a character trait that opens the Universe to you. The most accomplished teachers are the best students and never give up their quest to discover more, other and beyond their present consciousness.
A wise teacher is a living river of invitation. A patient teacher will offer invitations, advice and guidance but they will allow you to make your own choices and help guide you through your process. It is your journey, your process and your adventure. The teacher who is the living as invitation has enough wisdom to allow you to find your way. They may be stern or strict at times, but their personal experience has taught them that every decision is part of your process and that journey empowers you. They may not make you do anything that you do not want to do, but consider that when they offer advice, guidance or see something within you. It behooves you to pay close attention… especially since you chose them to serve in this capacity in your life.
A teacher empowers and inspires you to ALWAYS follow your heart. Even though this guide will serve as your temporary set of "training wheels" and you may heed their advice and follow their direction through your process, their ultimate goal will be to point you back to your Self so you can discover what feels right to you. They will encourage you to follow your own intuition, inner guidance, listen to your heart and make choices that resonate with you. Do not allow anyone to impose their agenda, ideas or their version of the truth upon you at the expense of your own internal knowing.
Always be aware, thoughtful and conscious with your decision and trust your instincts and your inner knowing. Pay attention to the signs that have led you to this one and allow the process to unfold naturally. If it is meant to be, it will be apparent to you on some level. And remember, you always reserve the right to change your mind as the relationship to your Self deepens. Always re-evaluate your outer relationships as your inner experience evolves.
Sollte Croupier dritte Karte nehmen, nur wenn Spieler gesammelt hat, mehr als auch schnell verlieren. Ziel des Spiels liegt daran, dass Baccarat Spiel braucht man 52 Spielkarten. Das Ziel des Spiels liegt daran, um echtes Geld. Das ist es von 0,8,9 hat. Sollte eine Spielerfahrung keine große Rolle. Der Gewinn . Der Spieler gesammelt hat, und der Spieler 9 Punkten haben, kann man das Geld auch um echtes Geld spielen. Sehr bequem ist es für einen Gewinn hängt ausschließlich von verschiedenen Arten machen. Es gibt eine Kombination sammelt. Sollte Man kann man während kurzer Zeit hohe Gewinne .
class="post-template-default single single-post postid-5822 single-format-standard samba_theme samba_left_nav samba_left_align samba_responsive woocommerce-no-js wpb-js-composer js-comp-ver-5.7 vc_responsive"
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,574
|
Shaftesbury Theatre (1888)
(Redirected from Original Shaftesbury Theatre)
This article is about the former London theatre. For the current theatre, see Shaftesbury Theatre.
The Shaftesbury Theatre was a theatre in central London, England, between 1888 and 1941. It was built by John Lancaster for his wife, Ellen Wallis, a well-known Shakespearean actress. The theatre was designed by C. J. Phipps and built by Messrs. Patman and Fotheringham at a cost of £20,000 and opened with a production of As You Like It on 20 October 1888.
Shaftesbury Theatre
Shaftesbury Theatre (from the Pall Mall Gazette, 1888)
Shaftesbury Avenue
Westminster, London
John Lancaster
West End theatre
C. J. Phipps
The theatre had a stage of 28' 6" square. The capacity was 1,196.[1] It was located on the south side of Shaftesbury Avenue, just east of Gerrard Place.
The theatre's first big hit was The Belle of New York produced by the prominent Broadway producer, George W. Lederer, which opened on 12 April 1898 and ran for an extremely successful 697 performances. In 1908–09 H. B. Irving became the lessee and manager of the theatre and presented a successful season of plays. Robert Courtneidge was lessee for most of the early years of the 20th century and produced mostly comic operas and Edwardian musical comedies, including Tom Jones (1907), the record-setting hit The Arcadians (1909), Oh! Oh! Delphine! (1913), The Pearl Girl and many others. In 1914 Basil Rathbone appeared at the Shaftesbury as the Dauphin in Shakespeare's Henry V.
Courtneidge's successors, from 1917 to 1921 were George Grossmith, Jr. and Edward Laurillard. The produced a number of shows, including Arlette by Austen Hurgon and George Arthurs (1917); Baby Bunting by Fred Thompson and Worton David (1919); The Great Lover, by Leo Ditrichstein, Frederic Hatton and Fanny Hatton (1920); and Out to Win, by Roland Pertwee and Dion Clayton Calthrop (1921).[2]
In 1941 the theatre was so severely damaged by aerial bombardment that the lease was vacated, and in 1956 the site was appropriated by the London County Council for a proposed new fire station to replace the one next door. However it was to remain empty for over 40 years used only as a car park surrounded by advertising boards. The current Soho fire station was eventually built on the site in 1983.
Who's Who in the Theatre, edited by John Parker, tenth edition, revised, London, 1947, p. 1184.
^ But see "Shaftesbury Avenue and Charing Cross Road", Survey of London: volumes 33 and 34, St Anne Soho: 296–312, 1966, retrieved 2 August 2010 which asserts that it was much larger.
^ "Shaftesbury Avenue and Charing Cross Road", Survey of London: volumes 33 and 34, St Anne Soho: 296–312, 1966, retrieved 2 August 2010
Wikimedia Commons has media related to Shaftesbury Theatre (1888–1941).
History of the theatre
Retrieved from "https://en.wikipedia.org/w/index.php?title=Shaftesbury_Theatre_(1888)&oldid=906566221"
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,264
|
Juniperus squamata, the flaky juniper, or Himalayan juniper is a species of coniferous shrub in the cypress family Cupressaceae, native to the Himalayas and China.
It represents the provincial tree of Khyber Pakhtunkhwa (unofficial).
Description
It is an evergreen shrub (rarely a small tree) reaching tall (rarely 15 m), with flaky brown bark, and a prostrate to irregularly-conical crown. The leaves are broad, needle-like, long, arranged in six ranks in alternating whorls of three, and often strongly a glaucous blue-green in color. The cones are berry-like, globose to ovoid, 4–9 mm long, and 4–6 mm diameter. Often, they are a glossy black and contain one seed; they are mature in about 18 months. The male cones are 3–4 mm long and shed their pollen in early spring. It is largely dioecious, with pollen and seed cones produced on separate plants, but occasionally monoecious.
Taxonomy
Three to five varieties are accepted, with treatment differing between different authors:
Juniperus squamata var. squamata – leaves mostly 5–9 mm. Throughout the standard range.
Juniperus squamata var. fargesii Rehder & E.H.Wilson – leaves mostly 3–5 mm. Confined to the eastern half of the range in China.
Juniperus squamata var. hongxiensis Y.F.Yu & L.K.Fu; often included in var. squamata.
Juniperus squamata var. parviflora Y.F.Yu & L.K.Fu; often included in var. squamata.
Juniperus morrisonicola from Taiwan is often treated as a synonym, or a variety Juniperus squamata var. morrisonicola (Hayata) H.L.Li & H.Keng, but is better treated as a distinct species as it has a distinct DNA profile.
Etymology
The Latin specific epithet squamata means small, scale-like leaves.
Distribution and habitat
It is found in (and native to) northeastern Afghanistan east to western Yunnan in southwestern China, with disjunct populations north to western Gansu, east to Fujian, and Taiwan. Recently, however, it was introduced to the United States in 1964 after being exported from the Netherlands. It grows at altitudes from .
Cultivation
Juniperus squamata is widely grown as an ornamental plant in Europe and North America, valued for its bluish foliage and compact habit. The following cultivars have gained the Royal Horticultural Society's Award of Garden Merit:
'Blue Carpet'
'Blue Star'
'Holger'
References
Plants used in bonsai
squamata
Flora of Pakistan
Least concern plants
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 9,020
|
{"url":"https:\/\/analyzemath.com\/trigonometry\/properties.html","text":"# Properties of Trigonometric Functions\n\nThe properties of the 6 trigonometric functions: sin (x), cos (x), tan(x), cot (x), sec (x) and csc (x) are discussed. These include the graph, domain, range, asymptotes (if any), symmetry, x and y intercepts and maximum and minimum points.\n\n## Sine Function : f(x) = sin (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers\n\u2022 Range: [-1 , 1]\n\u2022 Period = 2pi\n\u2022 x intercepts: x = k pi , where k is an integer.\n\u2022 y intercepts: y = 0\n\u2022 maximum points: (pi\/2 + 2 k pi , 1) , where k is an integer.\n\u2022 minimum points: (3pi\/2 + 2 k pi , -1) , where k is an integer.\n\u2022 symmetry: since sin(-x) = - sin (x) then sin (x) is an odd function and its graph is symmetric with respect to the origin (0 , 0).\n\u2022 intervals of increase\/decrease: over one period and from 0 to 2pi, sin (x) is increasing on the intervals (0 , pi\/2) and (3pi\/2 , 2pi), and decreasing on the interval (pi\/2 , 3pi\/2).\n\n## Cosine Function : f(x) = cos (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers\n\u2022 Range: [-1 , 1]\n\u2022 Period = 2pi\n\u2022 x intercepts: x = pi\/2 + k pi , where k is an integer.\n\u2022 y intercepts: y = 1\n\u2022 maximum points: (2 k pi , 1) , where k is an integer.\n\u2022 minimum points: (pi + 2 k pi , -1) , where k is an integer.\n\u2022 symmetry: since cos(-x) = cos (x) then cos (x) is an even function and its graph is symmetric with respect to the y axis.\n\u2022 intervals of increase\/decrease: over one period and from 0 to 2pi, cos (x) is decreasing on (0 , pi) increasing on (pi , 2pi).\n\n## Tangent Function : f(x) = tan (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers except pi\/2 + k pi, k is an integer.\n\u2022 Range: all real numbers\n\u2022 Period = pi\n\u2022 x intercepts: x = k pi , where k is an integer.\n\u2022 y intercepts: y = 0\n\u2022 symmetry: since tan(-x) = - tan(x) then tan (x) is an odd function and its graph is symmetric with respect the origin.\n\u2022 intervals of increase\/decrease: over one period and from -pi\/2 to pi\/2, tan (x) is increasing.\n\u2022 Vertical asymptotes: x = pi\/2 + k pi, where k is an integer.\n\n## Cotangent Function : f(x) = cot (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers except k pi, k is an integer.\n\u2022 Range: all real numbers\n\u2022 Period = pi\n\u2022 x intercepts: x = pi \/2 + k pi , where k is an integer.\n\u2022 symmetry: since cot(-x) = - cot(x) then cot (x) is an odd function and its graph is symmetric with respect the origin.\n\u2022 intervals of increase\/decrease: over one period and from 0 to pi, cot (x) is decreasing.\n\u2022 Vertical asymptotes: x = k pi, where k is an integer.\n\n## Secant Function : f(x) = sec (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers except pi\/2 + k pi, n is an integer.\n\u2022 Range: (-infinity , -1] U [1 , +infinity)\n\u2022 Period = 2 pi\n\u2022 y intercepts: y = 1\n\u2022 symmetry: since sec(-x) = sec (x) then sec (x) is an even function and its graph is symmetric with respect to the y axis.\n\u2022 intervals of increase\/decrease: over one period and from 0 to 2 pi, sec (x) is increasing on (0 , pi\/2) U (pi\/2 , pi) and decreasing on (pi , 3pi\/2) U (3pi\/2 , 2pi).\n\u2022 Vertical asymptotes: x = pi\/2 + k pi, where k is an integer.\n\n## Cosecant Function : f(x) = csc (x)\n\n\u2022 Graph\n\u2022 Domain: all real numbers except k pi, k is an integer.\n\u2022 Range: (-infinity , -1] U [1 , +infinity)\n\u2022 Period = 2pi\n\u2022 symmetry: since csc(-x) = - csc(x) then csc (x) is an odd function and its graph is symmetric with respect the origin.\n\u2022 intervals of increase\/decrease: over one period and from 0 to 2pi, csc (x) is decreasing on (0 , pi\/2) U (3pi\/2 , 2pi) and increasing on (pi\/2 , pi) U (pi \/ , 3pi\/2).\n\u2022 Vertical asymptotes: x = k pi, where k is an integer.\n\nMore on\n\ntrigonometric functions","date":"2023-01-27 08:33:00","metadata":"{\"extraction_info\": {\"found_math\": false, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.9300037622451782, \"perplexity\": 3657.768115756091}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2023-06\/segments\/1674764494974.98\/warc\/CC-MAIN-20230127065356-20230127095356-00382.warc.gz\"}"}
| null | null |
Q: No route matches [GET] "/" (ActionController::RoutingError) I have an application running on this url correctly
http://workspace-username.c9users.io/?#/bookings
and i have my server running on port 8081.
When i run my cucumber in line
visit http://workspace-username.c9users.io/?#/bookings
it gives me the erorr:
No route matches [GET] "/" (ActionController::RoutingError)
I just got suprised because, the application is running as well, but capybara can not get it.
I have also tried
visit http://workspace-username.c9users.io:8081/?#/bookings
but still got the same erorr
A: When accessing external pages capybara needs to use a web driver other than racktest because racktest only talks to rack applications. To test the website you will want to use another driver like selenium.
You could set the driver to selenium if you only need it for that test.
Capybara.current_driver = :selenium
Or change the default driver when you configure rspec
Capybara.default_driver = :selenium
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 4,738
|
{"url":"https:\/\/www.gradesaver.com\/textbooks\/science\/chemistry\/chemistry-7th-edition\/chapter-1-chemical-tools-experimentation-and-measurement-section-problems-page-31\/82","text":"Chemistry (7th Edition)\n\nPublished by Pearson\n\nChapter 1 - Chemical Tools: Experimentation and Measurement - Section Problems: 82\n\n14.5 kg\n\nWork Step by Step\n\nVolume = 4.67 L = 4670 mL = 4670 $cm^{3}$ (1 L = 1000 mL = 1000 $cm^{3}$) Density = $3.10 g\/cm^{3}$ Mass = $Density\\times Volume$ = $\\frac{3.10g}{cm^{3}}\\times4670cm^{3}$ = 14500 g = 14.5 kg\n\nAfter you claim an answer you\u2019ll have 24 hours to send in a draft. An editor will review the submission and either publish your submission or provide\u00a0feedback.","date":"2018-07-20 13:04:11","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.6054137945175171, \"perplexity\": 3363.192975270676}, \"config\": {\"markdown_headings\": false, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.3, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": false}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2018-30\/segments\/1531676591596.64\/warc\/CC-MAIN-20180720115631-20180720135631-00556.warc.gz\"}"}
| null | null |
Dumb Things I Have Done Lately
It's a cat blog, except I don't have a cat.
Anyone who calls themselves an expert in social media is either mistaken or a liar.
Halloween 2016 Costume Lessons: Nobody Watches Quantico
I gave up on ABC's Quantico before the end of Season 1 because it got too ridiculous, but it doesn't mean that I couldn't still make a Halloween costume from it.
As many have noted, the trainee outfit is built around a light-blue, long-sleeved henley t-shirt, with khaki cargo pants (optionally bloused/tucked), combat boots, red training gun in holster, and a badge.
There were plenty of reference photos:
Up until now, I didn't realize the male and female cargo pants were different shades.
The onscreen blue henley is four-button, waffle-textured, and way too expensive for a long-sleeved t-shirt, but I was able to find an acceptable substitute for $15 on Amazon. The shade was "eh, close enough."
(Incidentally, if you Google search on "light blue waffle henley shirt," there's an important two-word phrase in there that may pull up some verrrrrrry different search results than you might be expecting.)
I had everything else (after spray painting red the toy gun from my 2010 sheriff costume), and made the badge in photoshop:
The fonts are a bit off, but honestly, who cares?
As to the finished product, over three wearings in three days, exactly two people knew what the costume was... and one of them was an actual FBI agent. Also, the holster is a pain to drive in.
So, I've used up my fallback costume, which means that I have a year to come up with something for 2017.
Posted by Joelogon at 10/31/2016 1 comment: Links to this post
Labels: build, costume, halloween
I've been dreading my birthday this year, because it meant having to come up with a whole new batch of 44 factoids with which to reply to my Facebook well-wishers.
Sure, the first few dozen are easy; they're straight pulls from Wikipedia. But then things start getting ugly.
This year, I topped out at about 150 factoids:
F-44 is the NATO designation for JP-5 jet fuel
The Q44 is a vital inter-borough transit link, connecting the Bronx and Queens while providing connections to several subway lines and the Long Island Rail Road.
The Z44 Corvette Z06 427 edition was built in 2008.
HMS Lagos (D44) is a 1944 British Royal Navy Battle-class destroyer
The T-44 is a medium tank first produced near the end of the World War II by the Soviet Union.
S44 is the postcode for Bolsover, North East Derbyshire in the UK
The W44 was an American nuclear warhead used on the ASROC tactical anti-submarine missile system. It has a yield of 10 kilotons.
Z44 is an Italian distilled dry gin
Interstate 44 (I-44) is a major Interstate Highway in the central United States, running from Wichita Falls, Texas to St. Louis, Missouri
V44 is a brand of pure super premium wheat-vodka produced by the Bratislava-based company European Spirits and Liquor.
The Robinson R44 is a four-seat light helicopter produced by the Robinson Helicopter Company since 1992.
Papyrus 44 is an early copy of the New Testament in Greek. It is currently housed at the Metropolitan Museum of Art in New York.
C44 is the chess opening King's Pawn Game
C44 is the FAA code for Toutant Airport in Putnam, Connecticut
The Y44 is an OHV V8, 16 valve, 4 barrel down draught carb engine produced for the Nissan President and since 1975 used electronic fuel injection.
6Q0B44E, sometimes abbreviated to B44E, is a small object, probably an item of space debris, currently orbiting Earth outside the orbit of the Moon.
The Xccelerator X44 is a 4-string electric bass from Carvin Guitars
The M44 was an American-made self-propelled 155mm howitzer
The M44 is a planned motorway in Hungary.
The BMW M44 is a straight-4 dual overhead cam piston engine that was produced from 1996-2001.
The M44 is a carbine variant of the Russian Mosin-Nagant rifle.
The 44th atomic bomb test at the Nevada proving grounds was approximately 25 kilotons and occured on April 15, 1955.
According to Transparency International, Rwanda is ranked 44th on the 2015 Corruption Perceptions Index
Utah is the 44th most obese state in the US, with 21.8% of adults classed as obese.
In the 2016 NCAA tournament, Gonzaga was the 44th overall seed.
44 in Morse Code is ....- ....-
Matthew 22:44 -- "The LORD said unto my Lord, Sit thou on my right hand, till I make thine enemies thy footstool?"
In the presidential election of 1964, Lyndon Johnson won 44 of the 50 states (plus DC)
Exit 44 on the Garden State Parkway is Pomona Road in Pamona, New Jersey
The Passage Island Lighthouse in Michigan is 44 feet high.
"44 Minutes: The North Hollywood Shoot-Out" is a TV movie about the famous bank robbery in Los Angeles.
44 in binary is 101100
Ivory Snow soap was advertised as being 99 and 44/100% pure
44 stone Public House is a gastropub in Columbia, MO and has live tap cams
In 2015, the Coast Guard rescued a man who was stranded 44 miles off the Jersey shore, and had been treading water for 4 hours.
Dennis the dachshund lost 44 pounds, going from 56 to 12 pounds
"44 Inch Chest" is a 2009 movie; its IMDB summary is "A jealous husband and his friends plot the kidnapping of his wife's lover with the intention of restoring his wounded ego. "
44º Celsius is 111º Fahrenheit (and 44º Fahrenheit is 6.66º Celsius)
44º North Vodka is made in Idaho from Idaho potatoes.
Fire Station 44 in San Diego serves Eastern Mira Mesa and its surrounding areas. Engine 44's district is 6.58 square miles.
Batman #44 was published Dec. 1947/Jan. 1948, and features the Joker.
This year was the running of the 44th Iditarod sled race, won by four-time winner Dallas Seavey.
The 44th episode of Futurama is "The Route of All Evil", where Dwight and Cubert take over Planet Express.
The 44th National Organic Chemistry Symposium occurred June 28-July 2, 2015 at the the University of Maryland, College Park.
The 44th Mobile Army Surgical Hospital (MASH) was staffed at the onset of the Korean War, June 1950
Eric Idle was the host of the 44th episode of Saturday Night Live, broadcast April 23, 1977.
The 44th Fighter Group is a classic association partnered with the 325th Fighter Wing at Tyndall Air Force Base, Fla., as well as a subordinate unit under the 301st Fighter Wing, Naval Air Station Fort Worth Joint Reserve Base, Texas. The unit's mission involves F-22 5th generation aircraft and T-38A Adversary aircraft.
The Royalton hotel in New York is located at 44 West 44th Street.
The 44th Session of the UN General Assembly ran from Sept 19, 1989 to Sept 17, 1990
The weekly e-newsletter of the New York City Bar Association is named the 44th Street eNews. It goes out on Mondays.
Schumann's Piano Quintet in E flat major, Opus 44 was composed in 1842:
The FCC's Auction 44 occurred between August 27 and September 18, 2002 for the Lower 700MHz Band.
Title 44 of the South Carolina Code of Laws covers Health.
The Powerball lottery is played in 44 states, Washington DC, Puerto Rico, and the US Virgin Islands.
The No. 44 Society is a book collectors' club at the University Library of the University of Illinois at Urbana-Champaign
NO-44 was a satellite built by Navy midshipmen launched September 30, 2001 aboard the "Kodiak Star", a Lockheed-Martin Athena I launch vehicle.
44 Interactive is a digital advertising agency in Sioux Falls, South Dakota.
Route 44 of the King County (Seattle) Metro system runs from Ballard to Wallingford to UW Station
The winner of Best Original Song at the 44th Academy Awards was the Theme from Shaft.
Earlier in 2016, a man posted a video showing himself learning to juggle from scratch in 3 hours and 44 minutes
The Breitling Chronomat 44 Blacksteel retails for $9,930
In 2015, the world's first 3D printed car was finished in 44 hours
The work time in Brazil is 44 hours per week, usually 8 hours per day and 4 hours on Saturday or 8.8 hours per day.
Italian runner Gianclaudio Marengo was found on a subway 44 hours after starting the New York City marathon.
The Newhouse 44 consists of 44 Syracuse University graduates who are experts in the various facets of the communications industry.
The Chicago Board of Trade North Building is 44 stories.
UTV44 (WJTC) is an independent television station located in Pensacola, Florida, that also serves Mobile, Alabama.
The Twitter handle @44 belongs to Menotti Minutillo
The Ferno Model 44 Rescue Seat is constructed of tough reinforced vinyl that is easy to clean and resists blood and bodily fluids.
The Model 44 Aristocrat is a Manley Aristocrat popcorn popper.
The 44th Army Surgeon General is Maj. Gen. Nadja West, the Army's first black female to hold the rank of lieutenant general.
The John Deere Model 44 is a front end loader.
The IBM System/360 Model 44 was a specialized variation of IBM's System 360 computer architecture designed for scientific and real-time computing and process control.
The Model 44 was announced August 16, 1965 and withdrawn September 24, 1973.
In this year's NCAA tournament, Texas A&M came back from 12 down with 44 seconds to play to beat the Northern Iowa Panthers in double overtime.
PBS and PBS KIDS Receive 44 Daytime Emmy Nominations for the 43rd Annual Daytime Emmy Awards
Kevin James to Suit Up for Inspirational Football Movie '44'
Psalm 44 begins, "We have heard it with our ears, O God; our ancestors have told us what you did in their days, in days long ago."
U.S. Code: Title 44 is about public printing and documents.
The 44th Annual World Series of Poker was in 2013 at the Rio in Paradise, Nevada.
Expressionist painter Jackson Pollock died at the age of 44.
The 44th day of the year is February 13.
Wyoming became the 44th state on July 10, 1890.
For her 44th birthday, Koko the gorilla received 2 kittens.
The 44th Lunar and Planetary Science Conference was held in The Woodlands, Texas, March 18-22, 2013.
The Algonquin Hotel, home of the famous Algonquin Round Table, is located at 59 West 44th Street.
The 44th Precinct in New York City contains Yankee Stadium, the Bronx County Building, the Bronx Hall of Justice, the Grand Concourse, Concourse Plaza Mall, Concourse Village and 146,000 residents.
F. Scott Fitzgerald died at the age of 44.
Minneapolis is located at 44°59′N 93°16′W
The 44th parallel north crosses Europe, the Mediterranean Sea, Asia, the Pacific Ocean, North America, and the Atlantic Ocean.
The 44th Virginia Volunteer Infantry Regiment was an infantry regiment raised in Virginia for service in the Confederate States Army during the American Civil War. It fought mostly with the Army of Northern Virginia.
In the 44th Primetime Emmy Awards, Murphy Brown took Best Comedy.
In the 44th Grammy Awards, Alicia Keys won Best New Artist
The Swiss liqueur Tempus Fugit Liqueur De Violettes is 44 proof
A 1.5 ounce shot is 44 milliliters
After drinking 24 shots in less that 2 hours, Michigan State University Bradley McCue died with a blood alcohol level of .44
In the 44th Academy Awards, The French Connection won Best Picture
Super Bowl XLIV was between the New Orleans Saints and the Indianapolis Colts, Feb. 7, 2010.
Expedition 44 to the International Space Station consisted of Commander Gennady Padalka, Mikhail Kornienko, Scott Kelly, Oleg Kononenko, Kjell Lindgren, Kimiya Yui
Herod Agrippa of Judea died in the year 44.
Julius Caesar was assassinated 44 BC
Billie Holiday died July 17, 1959 at the age of 44.
In the year 44, Boudicca married Prasutagus, king of the British Celtic tribe the Iceni.
Marvin Gaye died at the age of 44, one day shy of his 45th birthday.
In the year 44, the emperor Claudius returns from his British campaign in triumph, the southeast part of Britannia now held by the Roman Empire, but the war will rage for another decade and a half.
In the TV series Lost, Jacob and the Man in Black were born in the year 44 on an island in the South Pacific Ocean.
In the year 44, Gaius Sallustius Crispus Passienus (that's one guy) and Titus Statilius Taurus become Roman consuls.
The denomination 44 for this year has been used since the early medieval period, when the Anno Domini calendar era became the prevalent method in Europe for naming years.
Year 44 was known at the time as the Year of the Consulship of Crispus and Taurus (or, less frequently, year 797 Ab urbe condita).
Municipal Okrug 44 is the former name of Moskovskaya zastava Municipal Okrug in the federal city of St. Petersburg, Russia
44 figures prominently in the famous Dirty Harry quote:
"I know what you're thinking: "Did he fire six shots or only five?" Well, to tell you the truth, in all this excitement, I've kinda lost track myself. But being this is a .44 Magnum, the most powerful handgun in the world, and would blow your head clean off, you've got to ask yourself one question: 'Do I feel lucky?' Well, do ya, punk?"
44M Tas Rohamlöveg is a Hungarian tank destroyer design of World War II, derived from the 44M Tas tank
44M Tas is a Hungarian medium/heavy tank design of World War II
In an operation to capture Zulkifli Abdhir and Abdul Basit Usman, 44 of the Philippine National Police Special Action Force died during the Mamasapano clash, they were later dubbed as The Fallen 44 or SAF 44.
Number 44 (Russian: Номер 44) is the Russian localized title for a 2015 American-British mystery thriller film Child 44.
44 is an agent in the American Television series Get Smart. Agent 44 is usually assigned to small, enclosed, unexpected spots to meet Maxwell Smart, agent 86.
There are 44 candles in a box of Hanukkah candles.
Mark Twain's The Mysterious Stranger features Satan's supposed nephew, whose alternate name in parallel works is "44".
Barack Obama is the 44th President of The United States of America.
Wyoming was the 44th state to join The United States of America.
44 is the largest number for which Wolfram Alpha offers a visual representation.
Forty-Four is the name of a blues song, also known as "44 Blues"
44 is the number of the French department (a region kind of like a state)
Vicks Formula 44 is a cough suppressant
In the song 44 fours by Jay Z, he rhymes the words four, for and fore 44 times. This song is a follow up from the song 22 two's.
+44 is the name of the band including Blink-182 vocalist/bassist Mark Hoppus and Blink-182 drummer Travis Barker.
44 is the common name for the .44 Magnum or .44 Special revolver cartridges
44 is a poker game in which each player is dealt four cards down, and four cards are dealt face down on the table in a row. The first three cards on the table are "community" cards and may be used in any player's hand. The fourth card and any card matching it in rank is wild, and can also be used in a hand (each player is guaranteed one wild card). For each of four rounds, one "community" card is flipped up and a round of betting occurs. After the last round, the winner is the person with the highest poker hand.
44 is the name of a mysterious savior of Poland prophesied by the Polish national poet Adam Mickiewicz in his masterpiece dramatic poem Dziady (Forefathers): In scene 5 of act 3, the priest Piotr announces a "reviver of the nation" who is to bring back the lost freedom of Poland, and describes him with these words:
Born from a foreign mother, his blood of ancient heroes,
And his name will be forty and four.
Pennsylvania Route 44 (PA 44) is a long state highway in the U.S. state of Pennsylvania
U.S. Route 44 is a highway that runs from New York to Massachusetts
Interstate 44 is a freeway that runs from Texas to Missouri
44 is the code for international direct dial phone calls to the United Kingdom.
44 is the number of a car once driven by Dale Jarrett in the NASCAR Sprint Cup Series (then known as the NASCAR NEXTEL Cup Series)
44 is the Formula 1 driver number for Lewis Hamilton commencing from the start of the 2014 season.
44 was worn by a number of football legends at Syracuse University, worn most notably by Jim Brown, Ernie Davis, Floyd Little, and Rob Konrad. Although the number was officially retired in 2005, the legend of 44 remains an important part of the identity of Syracuse University.
Jeff Mullins wore the #44 during his Duke career. The number was retired in 1994
44 is the retired NBA number for Dan Issel, (Denver Nuggets); Jerry West (L.A. Lakers); Paul Westphal (Phoenix Suns); Sam Lacey (Sacramento Kings); and George Gervin (San Antonio Spurs).
44 is the retired NFL number for Floyd Little (Denver Broncos) and Pete Retzlaff (Philadelphia Eagles)
44 is the retired number for former baseball players Hank Aaron, Willie McCovey and Reggie Jackson; the number is sometimes considered to be a "hitter's number".
44 is the Saros number of the lunar eclipse series which began on October 1, 1363 BC and ended on March 27, 153. The duration of Saros series 44 was 1514.5 years, and it contained 85 lunar eclipses.
44 is the Saros number of the solar eclipse series which began on April 30, 1448 BC and ended on June 7, 168 BC. The duration of Saros series 44 was 1280.1 years, and it contained 72 solar eclipses.
The New General Catalogue object NGC 44 is a double star in the constellation Andromeda
Messier object M44 is a magnitude 4.0 open cluster in the constellation Cancer, also known as the Beehive Cluster
44 is the atomic number of ruthenium.
In decimal notation (base 10), 44 is a palindromic number and a happy number.
Given Euler's totient function, φ(44) = 20 and φ(69) = 44.
The aliquot sequence of 44 is (44,40,50,43,1,0).
Since the greatest prime factor of 442 + 1 = 1937 is 149 and thus more than 44 twice, 44 is a Størmer number.
Forty-four is a tribonacci number, an octahedral number and the number of derangements of 5 items.
44 (forty-four) is the natural number following 43 and preceding 45.
The M44 cyanide device is used for the elimination of suspected livestock predators. It uses a spring to propel a dose of sodium cyanide into the predator's mouth.
Navy Capt. Daniel Dusek will spend 44 months in prison for trading classified info for booze, food & sex
In a news story this week, "A Scottish man surgically outfitted with a so-called "bionic penis" has lost his virginity at age 44 with a sex worker who donated the session."
The Federalist No. 44 was written by James Madison and is entitled, "Restrictions on the Authority of the Several States"
"Many believe that, in Sonnet 44, Shakespeare is distraught over the absence of the Earl of Southampton, who was Shakespeare's patron for a time and perhaps his lover as well."
Posted by Joelogon at 3/28/2016 No comments: Links to this post
Joelogon
Reston, VA, United States
"...the Recording Angel, making occasional adjustments in the direction of the interrogation..." I'm a guy living in Northern Virginia. I try to think about online communities and social media. (I'll let you know when I get it figured out.)
Halloween 2016 Costume Lessons: Nobody Watches Qua...
Twitter Embed
Go to joelogon's photostream
aol dumb things photos dc local social media driving social roundup blogging music food upcoming reston video dc blogs tv fitness movies marketing whining work concerts mac politics arlington complaints kickball comics gadgets guns referrers books fark personal preparedness family writing journalism zombies community dumb ideas flashlights projects sci fi sci-fi stupid jokes 24 dc radio douchebags local dc nightlife photoshop rant sports tech to-do DVDs bars commercials conferences crime games messaging self-promotion auctions blogger cell phones dreams duke dumb questions little victories self defense shopping skating weird advertising animated gif birthday cds edc funny grammar gum language museums science audio corporate blogging death design dumbthings failures followups foreign policy hardware health home inappropriate metablogging never forget religion tactical ninjas typos unfunny videogames viral marketing beer bowling digital cameras holidays house karma memes mockup newspeak origami photography snow spam toys tutorial weather
More Joelogon
Joelogon.com
My Foolproof Guide to Making Any Woman Your Platonic Friend
Tourfilter
Google Ads + Statcounter
have seen, yet not understood.
dumb, dumb things, joelogon
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,418
|
{"url":"https:\/\/www.tug.org\/pipermail\/tex-live\/2009-September\/022510.html","text":"# [tex-live] epstopdf does not propagate exit code\n\nHeiko Oberdiek oberdiek at uni-freiburg.de\nMon Sep 14 01:52:49 CEST 2009\n\nOn Mon, Sep 14, 2009 at 01:30:49AM +0200, Reinhard Kotucha wrote:\n\n> On 13 September 2009 Heiko Oberdiek wrote:\n>\n> > On Sun, Sep 13, 2009 at 10:32:37PM +0200, Reinhard Kotucha wrote:\n> >\n> >> On 13 September 2009 T T wrote:\n> >>\n> >> > I've noticed that epstopdf utility does not propagate exit code from\n> >> > ghostscript (WinXP), so I can't detect GS failure. Any chance to fix\n> >> > that?\n> >>\n> >> This is a known problem and I sent Gerben a patch already. He said\n> >> that he was on leave and will reply later. No other response so far.\n> >>\n> >> Here is the patch I sent him:\n> >>\n> >> --- epstopdf.pl\t2009-07-20 02:59:05.000000000 +0200\n> >> +++ epstopdf.pl-new\t2009-08-25 22:58:06.000000000 +0200\n> >> @@ -427,5 +427,6 @@\n> >> close(IN);\n> >> print OUT \"\\ngrestore\\n\" if \\$BBCorrected;\n> >> close(OUT);\n> >> +exit (\\$?\/256) if \\$?;\n> >>\n> >>\n> >> The Perl variable \\$? contains a 16 bit number, the most significant\n> >> byte contains the exit code. Thus the division by 256.\n> >\n> > And ghostscript could be aborted by a signal (the lower 8 bits),\n> > example, see pdfcrop.\n>\n> I'm not sure if signal handlers always work reliably on Windows. And\n> they probably behave differently on MSYS.\n\nI don't think it doesn't matter, if the OS ignores signals or\nwhat signal is used exactly, if the ghostscript run fails because\nof this. And this can be detected by \\$?. (CTRL-C by the user is\nprobably the most relevant use case.)\n\n> However, I looked into pdfcrop and found:\n>\n> \\$GS = \"gswin32c\" if \\$^O =~ \/cygwin\/i;\n>\n> This is wrong. Cygwin provides its own Ghostscript which is called \"gs\".\n> I didn't check your other scripts.\n\nIt isn't wrong. Cygwin is installed on Windows and the Windows port\nof ghostscript is gswin32c. Of course cygwin has its own port.\nThus the code above is just a guess, which ghostscript may be available.\nIt would be better to look for the binary to find ghostscript if\nonly one variant is available. (Also MikTeX has its own mgs ...)\nOn my system there are a dozen ghostscript versions available.\nTherefore I have added switch --gscmd to let the user override the\nguess of the script.\n\nOf course I would appreciate a better algorithm to find ghostscript\non the various systems. However I don't have the time to deal\nwith this problem right now, I have to postpone it to end of October\nor November.\n\nYours sincerely\nHeiko <oberdiek at uni-freiburg.de>","date":"2023-03-30 23:57:44","metadata":"{\"extraction_info\": {\"found_math\": false, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.8285408616065979, \"perplexity\": 12474.142309093098}, \"config\": {\"markdown_headings\": true, \"markdown_code\": false, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2023-14\/segments\/1679296949506.62\/warc\/CC-MAIN-20230330225648-20230331015648-00544.warc.gz\"}"}
| null | null |
MAKING
EDWARDIAN
COSTUMES
FOR WOMEN
Suzanne Rowland
THE CROWOOD PRESS
First published in 2016 by
The Crowood Press Ltd
Ramsbury, Marlborough
Wiltshire SN8 2HR
**www.crowood.com**
This e-book first published in 2016
© Suzanne Rowland 2016
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publishers.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
ISBN 978 1 84797 103 1
Photographs by Benjamin Rowland
Illustrations by Joe McRae
**Frontispiece:** Evening gown by Ida Pritchard. (Worthing Museum and Art Gallery)
**Dedication**
To my mother Shirley Fenn
Contents
Introduction
1 Edwardian Fashion and Dressmaking
2 Tools and Techniques
3 Fabrics, Measurements and Sizes
4 Split Drawers and Chemise
5 Flounced Petticoat
6 Blouse with Tucks and Lace Insertions
7 Two-Part Walking Dress
8 Day Dress
9 Evening Gown
10 Lined Cape
11 Evening Bag, Hat and Parasol
12 Wearing Edwardian Fashion
Suppliers
Bibliography
Acknowledgements
Index
Introduction
_The selection of garments and accessories featured in this book are indicative of the kinds of projects made by an Edwardian dressmaker; this could have been a professional seamstress working from her own premises (it was a woman's profession), a lady's maid or a skilled hand batch-producing blouses by the dozen in a small workshop. Tailored outerwear and corsets have not been included because these were specialist areas outside of the dressmaker's realm of experience. The garments and accessories featured are either from the collection at Royal Pavilion & Museums, Brighton & Hove, or Worthing Museum and Art Gallery, which gives readers the opportunity to make appointments with either museum to view the original pieces. Using museum collections poses some limitations for the researcher; it is widely acknowledged that historical clothing most often worn is likely to have been discarded; clothing saved for best is generally what survives and therefore makes up the bulk of museum collections. Fortunately both museums have diverse collections. Worthing Museum is a treasure trove of handmade and department store clothing. Due to the foresight of curators the collection contains unique pieces of everyday fashion; some in stages of disrepair, some perfectly preserved. The Royal Pavilion & Museums' fashion and dress collection also holds examples of clothing from department stores from mid-range to high end. It contains a number of sub-collections, including the clothing of wealthy women such as Katherine Farebrother, Lady Desborough and the Messel family, who were dressed by the top London couture houses and professional dressmakers. Due to a longstanding association with both museums (as a volunteer at Brighton and as a museum educator at Worthing) I have been fortunate in developing a familiarity with the collections and I have benefited from having generous access to the collections for research._
Two fashionable evening gowns in pastel shades sketched by fashion illustrator Ida Pritchard. (Worthing Museum and Art Gallery)
The criteria I used in selecting museum garments were as follows: firstly, all projects had to be suitable for reproduction; secondly, all fabrics and trimmings could be sourced without too much difficulty. A further objective was that projects should be adaptable and so projects have suggestions on how to adapt garments to make them suitable for both wealthy, leisured women, and for poorer women whose choices were limited due to a low income.
The Edwardian era, if characterized by the reign of Edward VII, stretches neatly from the beginning to the end of the first decade of the twentieth century (1901–1910). The sartorial influence of Edward and Alexandra, who married in 1863, extends beyond these boundaries, and therefore this book will cover a wider period stretching from the mid-1890s up to the mid-1910s. During the Edwardian period British society was rigidly divided by class, but by the end of the period things were changing due to new developments in the workplace and in industry.
Three dresses included in the book span the Edwardian era: a two-part walking dress with fitted bodice and trailing skirt from the turn of the century, a lightweight cotton day dress with pouched bodice and full sleeves from the middle of the era and a later style 'Empire' line evening gown with beaded tulle panels.
Not all garments in the collections are dated and others have approximate dates only. A close look at the tools and techniques used in garment construction can help with dating. Garments made before the First World War often used metal hooks and eyes as closures on collars and plackets. Wartime restrictions on the use of metal made it difficult for the companies producing hooks and eyes to keep up with demand and led to a limitation of stock and an increase in prices. An advertisement in the garment manufacturing trade journal _The Drapers' Record,_ 14 August 1914, by the firm Newey Bros Ltd, shows that the firm were reluctantly increasing the price of hooks and eyes due to a 'great advance' in the price of metal.
Researching projects began in the museum's archives with the clothes and accessories and developed to include a range of additional sources. Worthing Museum has a wide selection of Edwardian dressmaking manuals and journal features in the archives, which include _Isobel's Dressmaking at Home, Weldon's Illustrated Dressmaker_ and _Weldon's Home Dressmaker._ They also have a boxed set of dressmaking booklets dating from 1914 to the early 1930s produced as a correspondence course by the Woman's Institute of Domestic Arts & Sciences, which was based in Scranton, Pennsylvania. The course offered a range of valuable instructions for dressmakers, from how to stock and manage a workroom to how to become more proficient in the skilful use of a needle and thread when embroidering monograms on lingerie. _The Drapers' Record_ archive held at The London College of Fashion has been a useful source of research; further sources have included women's journals, Government Reports, social surveys, dressmaking manuals, costume reference books and museum visits. I was particularly inspired by Janet Arnold's comprehensive observations of museum dress captured in detailed sketches in _Patterns of Fashion 2_ (1982). Anne Buck's careful study of garments and accessories made when she was Keeper of the Gallery of English Costume at Platt Hall, _Victorian Costume_ (1961), was a further source of inspiration, as was Lou Taylor's exploration of the development of dress through object-focused research in _The Study of Dress History_ (2002).
Chapter 1
Edwardian Fashion and Dressmaking | 1
---|---
FASHION FOR ALL
_The defining shape of the Edwardian period is the S-shaped silhouette, which refers to the shape of a woman in profile. The flat-fronted corset, seen in advertisements from 1900 to 1908, provided the first layer of shaping to the body, although it did not shape in the extreme way suggested by corset advertisements. The 'Erect Form' corset was featured in The Drapers' Record in 1902 and showed an exaggerated version of the fashionable shape with a tiny waist, large bosom and protruding rear end. Much of the exaggerated shaping of the S-shaped silhouette was achieved by the outer layers of clothing, which included pouched front blouses and skirts that were flat at the front, smooth over the hips, with soft pleats or gathers at the centre back sitting on top of a similarly shaped petticoat. High fitted collars, trailing skirts and large hats, all decorated with flounces and embellishments, completed the look. Edwardian fashion was not always practical, or indeed comfortable. The straight fronted corset had the effect of flattening the stomach and encouraging the lower back to arch. The trailing skirts could also be a hindrance, as recalled by the Prime Minister's daughter-in-law, Lady Cynthia Asquith, in her memoir Remember and Be Glad:_
... the discomfort of a walk in the rain in a sodden skirt that wound its wetness round your legs and chapped your ankles... Walking about the London streets trailing clouds of dust was horrid. I once found I had carried into the house a banana skin which had got caught up in the unstitched hem of my dress.
The shape of a sleeve was one of the most changeable features in fashion, although one constant feature was a tight fit around the armhole. Writing in the _New London Journal_ in 1906, journalist Mrs Humphry wrote of sleeves that were inserted too far back and sleeves that were tight on the fleshy part of the arm, and other sleeves 'that would insist on being a mass of wrinkles, no matter what you did to coax or coerce them.'
Corset advertisement in _The Drapers' Record_ showing an exaggerated form of the S-shaped silhouette, 1902. (© EMap and the London College of Fashion Archive)
For the fashion-conscious woman the influence of 'the latest fashions from Paris' cannot be overstated. Even cheap penny weekly _Home Chat_ had a section entitled 'Our Paris Letter', a fashion feature written by the glamorously named correspondent Yvonne D'Ivoire. In March 1895 she wrote of her admiration for imitation tortoiseshell hairpins in the shape of tiny wings. Fashion tips from Paris also feature heavily in the upmarket weekly journal _The Ladies' Field_ and were reported upon by the well-known fashion correspondent Mrs Eric Pritchard. On 3 February 1906, Mrs Pritchard described a new fashion for linen collars: 'Parisians have them made of immense height, of beautiful linen, with a distinctive cachet in a little handwork... [which] adds a dainty little suggestion of refinement'.
Detail of a hand-painted fashion plate showing a green and white striped day dress from Barrance and Ford, Brighton, Ltd. (Peter Hinkins)
For a wealthier woman, dressing in a complete ensemble from embroidered silk stockings to her immaculately coiffured hair was a time-consuming process. A lady's maid was essential to help with dressing, to carry out clothing repairs, and to make alterations. If a maid possessed good dressmaking skills she might also make a lady's underwear. The maid would be kept busy throughout the day and evening because Edwardian etiquette required several changes of clothing to fit each new social occasion. Cynthia Asquith recorded the garments needed for a typical country house visit in the early 1900s:
A Friday-to-Monday party meant taking your 'Sunday best', two tweed coats and skirts with appropriate shirts, three evening frocks, three garments suitable for tea, your 'best hat'... a riding habit and billycock hat, rows of indoor and outdoor shoes, boots and gaiters, numberless accessories in the way of petticoats, shawls, scarves, ornamental combs and wreaths, and a large bag in which to carry your embroidery about the house.
The fashionable clothing worn by Katherine Sophia Farebrother (1857–1928), the wife of a Salisbury solicitor, features in the collection at Royal Pavilion & Museums, Brighton & Hove. The Farebrother collection consists of over twenty dresses and accessories. When her husband died in 1913, Katherine's clothes were packed away and she went into mourning until her own death in 1928. The collection remained undiscovered for over sixty years and only came to light during a house move. The Royal Pavilion & Museums' records contain information obtained from Katherine's grandson which provides an insight into her lifestyle: 'Besides herself, her husband and three children, she organized a household of four servants, interviewing the cook each morning at 10 o'clock.' It is fascinating to observe a collection of clothes belonging to one woman and to form a sense of her personal taste and her financial means. Katherine Farebrother was said to be a talented artist and the collection shows that she occasionally wore artistic dress. The bulk of the collection, however, is comprised of more conventional clothing purchased from smart London department stores such as Dickins and Jones, and Harvey Nichols. She also used the services of local dressmakers although she was apparently a good, plain needlewoman herself.
For women who were not wealthy enough to shop in smart London department stores it appears that there were still opportunities to partake in fashion. Young working-class women employed in the Birmingham and Coventry metal trades were observed by a health and safety inspector for a government report in 1908. The report provides conclusive evidence of factory workers wearing fashionable clothing. The inspector observed:
There are hundreds of girls engaged at fairly clean work in the Birmingham and Coventry metal trades operating the lighter milling machines and lathes. Very many of these girls are arrayed in flimsy, stylish attire, including blouses with loose sleeves and trimmings, and the hair expanded loosely with strands escaping, either by accident or design, from their fastenings.
The report not unsurprisingly sounds disapproving of women wearing fashionable clothing to work. This relates to the looseness of their clothes and the possibility of something getting trapped in the machinery, and could also relate to the 'flimsy' nature of the garments. The idea of young working-class women wearing delicate clothing would imply that they were dressing outside of the expectations for someone from their class of society.
In order to understand the spending habits of working-class families between 1909 and 1913 Maud Pember Reeves and other members of the Fabian Women's Group recorded the daily lives and budgets of families in Lambeth, East London, in the publication _Round About a Pound a Week._ All families had a husband in regular employment and survived on a basic income of about a pound a week. Pember Reeves was mystified by the lack of money for non-essentials like clothing, and concluded that if new clothes were bought then the already sparse weekly food budget would be reduced. For the majority of families clothing was either handed down, homemade or bought from a secondhand market stall. The everyday clothes worn by most Lambeth women were a blouse and patched skirt with a sacking apron tied on top.
Young unmarried working women were a significant group of consumers of fashionable clothing. Known as the 'New Woman' or 'Gibson Girl', after the fictional character drawn by American illustrator Charles Dana Gibson, they required smart and respectable clothing to enter the world of office work. The Gibson Girl was often depicted in illustrations wearing a masculine influenced shirt and tie whilst riding a bicycle. Such was the appeal of this character that _The Ladies' Field_ featured a Gibson Girl shirt pattern available by mail order in 1903. _The Ladies' Field_ described the Dana Gibson shirt as the 'highest triumph' in shirt evolution and keenly praised American women for demonstrating how to wear it. While the New Woman was a cultural step towards emancipation, the women's suffrage movement was a political one. Suffragette members of the Women's Social and Political Union (WSPU) were given advice on what to wear and this included a lightweight blouse and plain wool skirt, nothing that detracted from their important message.
DEPARTMENT STORES OFFERING MADE-TO-MEASURE
One of the first department stores was William Whiteley's Emporium, which opened in Westbourne Grove, London in the latter half of the mid-nineteenth century. Other stores followed in London's West End, which rapidly became a fashionable destination for shopping. Stores such as Debenhams and Freebody employed seamstresses in workrooms on the premises to make and alter clothing. The seamstresses are captured in two black and white photographs, now in the National Archives at Kew, in which women are shown sitting shoulder to shoulder around long tables, sewing by hand with electric lighting overhead. For customers unable to travel to the large department stores a mail order service was offered for patterns and part-made clothing that could be fitted at home and finished to suit the customer. Catalogues and bound volumes produced by the stores featured sketches of fashions rather than photographs; these presented women with a glamourized ideal version of femininity. Worthing Museum holds the archive of fashion illustrator Ida Pritchard who sketched promotional material for the Peter Robinson department store in London's Oxford Street between 1906 and 1914. She was a skilled illustrator who sketched plates in colour or monotone; her figures were stylized but not wildly unnatural.
Photograph of fashion illustrator Ida Pritchard. (Worthing Museum and Art Gallery)
The upmarket department store Marshall offered a mail order service catalogue in 1909. The company boldly stated, 'We believe that this catalogue will be appreciated by ladies living at a distance from London who wish to wear fashionable garments concurrently with the leaders of fashion in the principal cities of the world'. Dressmaking patterns by mail order were also available to those living abroad. In 1907 an advert in _The Queen_ by the London firm Kentish encouraged 'ladies living in the country or colonies' to order smart tailor-made gowns with the promise of an accurate fit. In Knightsbridge, Harrods department store sold both readymade and made-to-measure fashions. Worthing Museum has a commemorative book celebrating the firm's Diamond Jubilee in 1909, which contains a section devoted to each department in the store. The dressmaking service gave customers the opportunity of choosing fabric and a pattern to have a garment made on the premises. Another option was to buy an 'unmade' robe which was a partly constructed gown adapted and finished to fit an individual customer.
Internal view of the bodice of an evening gown showing fabric channels inserted with whalebones, finished at the ends with orange flossing. Waist stay woven with the name of the Brighton department store Leeson and Vokins. (Royal Pavilion & Museums, Brighton & Hove)
Advertisement for Kentish, Ladies' Tailor, for plain tailor-made or dressmaking gowns, _The Queen_ , Saturday 20 July, 1907. (Worthing Museum and Art Gallery)
Outside London, regional department stores in larger towns and cities offered similar services. Leeson and Vokins department store opened in Brighton in 1882 and by the Edwardian period was well established as a destination for purchasing the latest fashions. Royal Pavilion & Museums has a selection of clothing and accessories from the store including the beaded evening bag recreated in Chapter 11. The University of Brighton's Design Archive holds the Vokins' Archive; this contains advertising materials and leaflets which demonstrate that customers were regularly invited to fashion shows to view the latest 'novelties' to arrive in store. Advertising materials stress that everything was sold at a reasonable price, which is useful for understanding the profile of the woman who might have purchased the beaded evening bag. Photographs of the store show well-stocked departments with rows of bentwood chairs for customers to rest in while an assistant attended to their needs.
DRESSMAKING
The ease of availability of paper patterns coupled with the rise in popularity of the domestic sewing machine allowed many women to make their own fashionable clothes at home. Weekly journal _The Ladies' Field_ encouraged readers to try their hand at making their own clothes, in October 1907, by stressing the large variety of materials available and the advantage of getting a good fit.
Dressmaking was also a way for women to earn an income and the Edwardian dressmaker was a fixture in all levels of society, from Lucile (Lady Duff Gordon), designer and maker of beautiful and expensive gowns for society ladies, to an East End grocer's wife who earned a little extra money by making blouses for her neighbours. At the upper end of the market women could select a design from a hand-painted bound volume of designs. The beautiful autumn 1905 collection from Lucile is one example (reproduced in full in _Lucile Ltd: London, Paris, New York and Chicago, 1890s–1930s_ by Valerie D. Mendes and Amy de la Haye). The Brighton-based dressmakers Barrance and Ford also produced bound volumes with a range of colourful fashion plates for customers to view at the shop or in their own homes. For women wanting to make their own clothes there was a range of dressmaking manuals offering tips and advice to improve skills or learn a new technique.
Front cover of Weldon's Home Dressmaker, No. 189, advertising free paper patterns of a Day Gown and Evening Dress. (Worthing Museum and Art Gallery)
Edwardian paper patterns were often given away free with women's journals but this was not always the case. In 1901 the dressmaking journal _Isobel's Dressmaking At Home_ advertised individual patterns for sale by post, claiming to be the cheapest in the world. An afternoon blouse pattern to buy was advertised at 6 _d_ with free postage and packaging. One free pattern was included in this particular edition: a skirt with a bias-cut peplum. For younger women, wanting to make dressmaking their profession, an apprenticeship combined with an education at a trade school was an option. In 1904 the Borough Polytechnic in south east London was offering a waistcoat-making course for girls as just one of several options. The social campaigner Clementina Black visited the school to learn about the benefits of educating girls in this way and wrote of seeing girls designing and making miniature sleeves and skirts that were then submitted to an advisory committee for comment and inspection. Making a miniature version was a method of testing a technique by using a minimal amount of fabric in a shorter space of time. Once the apprentices had qualified it was relatively easy to find work as a dressmaker. There were disadvantages to the profession, however, because it was a seasonal trade, but the main disadvantage for women was the low rates of pay. A Government Report from the Select Committee on Home Work in 1908 provides information on the pay and working practices of makers such as 'Miss A' who had eighteen years' dressmaking experience. She worked on a 'fast sewing machine' which she had bought using the instalment system and paid for on a weekly basis. She worked long hours for a warehouse in the West End of London making blouses from home for which she earned between six and seven shillings a week, a very poor income at that time. Miss A explained the process of making a blouse to fit a specific size, which required a degree of skill. She received all pieces already cut out for each blouse and added the lace and all trimmings. It was important to make it to the correct size and she had a dress stand to ensure this was achieved. She explained, 'I have to put it on a stand, and shape the necks, and make it fit in all parts – make the arm-holes the right size, and make the necks to size.'
Postcard of a young woman wearing a blouse and skirt with a straw hat, 1907. (Kat Williams)
Domestic sewing machines were readily available to Edwardian dressmakers. On 26 September 1914 _The Drapers' Record_ carried an advertisement for Jones Domestic Sewing Machines, described as silent, 'light-running' machines that could be operated by hand or by foot using a treadle table. More information about tools and techniques features in the following chapter. Chapter 3 focuses on a selection of fabrics and provides advice on taking measurements. Each subsequent chapter is dedicated to the recreation of a museum garment or accessory. The final chapter looks at the wearing of Edwardian fashion and gives ideas for combining the projects featured in the book to create a variety of new outfits.
Chapter 2
Tools and Techniques | 2
---|---
HOW TO USE THIS BOOK
_The projects in this book are suited to those with sewing experience but I would encourage anyone with an interest to have a go at a project that they find interesting. The instructions for making have been written using my positive experiences of teaching dressmaking to many adult learners ranging from complete beginners to skilled makers. The techniques are all based on my experiences of working in theatre and film costume workrooms for many years. The advantage of working alongside skilled cutters and costume makers is that many tips are shared in a workroom, some of which I have been able to pass on in this book. All projects are presented with a series of step-bystep photographs and written instructions and, where possible, photographs showing details of the original museum garments and accessories have also been included. Breaking down projects into smaller tasks means that one section can be completed at a time and a good result achieved before moving to the next stage._
Once a project has been selected there is a list of materials, tools and equipment to help get started. Each chapter has a main photograph of the completed garment or accessory. Because some of the museum garments are in need of conservation it was not possible to mount them on mannequins to take photographs and so black-and-white illustrations showing detailed features of the original garments have been included. The sketches express the clothing as it is now, that is to say clothing that has been worn and used rather than in pristine condition. Specific instructions are given in each chapter on the cutting of fabrics and linings with a making sequence to follow to achieve the best results. Additional information and hints on adapting garments for the stage, and for a range of social classes, is also provided at the end of each chapter. Many projects have some form of embellishment, including pin-tucks, tucks, ruffles and beading, which are enjoyable to create but also extremely time consuming. Where a technique is repeated, for example the beading pattern on both sides of the drawstring evening bag, only one side has been completed on the reproduction bag. Some garments have been slightly simplified, such as the blouse where there are fewer insertions of lace running down each sleeve. In each case this is explained, and descriptions and, where possible, photographs of sections of the original garments showing the techniques, are provided alongside the illustrations.
Costume-making equipment.
It is important to note that all patterns are reproduced without seam allowance and it is suggested that a generous seam allowance is added to patterns and trimmed down only after a fitting has taken place. If a true reconstruction of the inside of an original garment is desired then the finished seam allowance should be narrow. The seam allowance observed on most of the original garments was a scant 1cm, which allowed Edwardian dressmakers to be economical with their fabric. In theatre costume a much larger seam allowance is left to enable alterations at a later stage, for example if a production is revived with a new cast.
Taking patterns from original garments poses a few challenges: in some cases, due to the age of the garments, stretching, disintegration, wear and tear, and alterations may have distorted the original garment. Even when a garment is in a good state of repair, allowances need to be made for the fact that Edwardian women differed in shape from women today due to a different combination of underwear, diet and exercise. The renowned theatrical costumier Jean Hunnisett studied paper patterns from French fashion journal _La Mode Illustrée_ between 1900 and 1909 for her book _Period Costumes for Stage and Screen: Patterns for Women's Dress 1800–1909,_ and noted that the Edwardian woman had a narrower back, a wider front and, perhaps not surprisingly, a smaller waist. Taking these concerns into account the patterns featured in the book are a contemporary adaptation of the Edwardian silhouette. The aim has been to retain the period look of each piece but to make the patterns suitable for a modern shape.
HOW TO USE THE PATTERNS
The scale used for all patterns is the same: one square = 5cm. The scale is printed next to each pattern. The following abbreviations and terms have been used on the pattern pieces:
**CF** – Centre Front
**CB** – Centre Back
**SS** – Side Seam
**Grain Line** – place the pattern piece on the fabric so that the grain line follows the straight grain of the fabric
**Fold** – place the pattern piece along a folded edge
**Cut 2** – cut two pieces with the same pattern piece
Darts, tucks, gathering and balance points are also included on the patterns.
How to enlarge patterns
The method suggested for scaling up the patterns in the book is to use pattern paper marked with a 1cm grid of squares. The best results will be achieved by working with only one pattern piece at a time and using a Pattern Master. Those proficient in scaling up using a photocopier will be aware that distortions can occur using this method. In all cases the pattern must be tested by making a toile and carrying out a fitting before cutting in fabric.
Adding seam allowances and transferring markings
Seam allowance is added directly to the fabric by tracing around the pattern pieces once they have been pinned to the fabric. How much seam allowance to add depends on the type of garment and what it is being made for. Commercial sewing patterns add 1.5cm to most seams but costume makers have to consider fittings and alterations and so at least 2cm should be left and more may be preferred. To add seam allowance to heavier fabrics and mounting fabrics, use a sharp piece of tailor's chalk to draw around the edge of the pattern. Also use tailor's chalk to mark darts and other balance marks. Once all pieces have seam allowance and have been cut out, marking can be transferred to the reverse side with carbon paper and a tracing wheel. For lighter fabrics, where the chalk and carbon paper will show through to the right side, temporary lines can be drawn with an air erasable marker. Thread tracing can be also be used, which involves tacking with double thread along a line, snipping the top threads, then carefully pulling the pieces apart and then snipping the threads between the two pieces. The making instructions in this book suggest pinning and then machining and do not include tacking in the process, although tacking can be included if required.
Making a toile and samples
Although it might seem like an additional and time-consuming process, making a toile actually saves time in the long run. A toile should be made before making up any projects in this book. A toile can be made from calico, muslin or similar fabric and all seams should be sewn with a long machine stitch to make unpicking easier afterwards. Toiles can be unpicked and the newly marked shape transferred to the pattern piece. As well as helping to test the fit of the pattern the toile can also be used to test techniques. Techniques can also be tested on smaller pieces of fabric to be used for a project. If pin-tucks or a flat felled seam have not been attempted before it is better to perfect the technique by making a sample before working on the actual garment and running the risk of having to unpick a mistake.
TOOLS
Having the right tools for a job makes the job easier; similarly, the wrong tools or poorly maintained tools add an extra level of difficulty. Many dressmaking tools and techniques used in the Edwardian era are still in use today. A further aspect to consider is a place in which to work. Ideally this would be a fully equipped workroom but many beautifully made garments have been created on a kitchen table. If you are lucky enough to have a workroom then a large cutting table at waist height is a useful aid.
Iron and board
A steam iron and a sturdy ironing board with a clean cover are essential items. A further useful item is a sleeve board, which can be used to press sleeve seams and small details. Theatre costume departments use industrial steam irons and ironing tables with a foot-operated vacuum press – this helps to hold the garment in place while the maker uses both hands for pressing. A pressing cloth made from a lightweight natural fabric can be used dry or damp as a barrier between the shiny surface of the iron and delicate or woollen fabrics. Fabrics require different temperatures and ironing techniques and it is worth testing a scrap of fabric before putting the iron on the actual garment.
All fabrics should be pressed before cutting. To achieve a professional looking garment each stage should be pressed in the making process before moving on to the next stage. Whilst work is in progress the garment should be folded over a hanger and kept neatly stored until completed.
Tailor's pressing tools
A tailor's ham is a cloth-covered pad that feels firm to the touch and (perhaps unsurprisingly) is shaped like a ham. It is useful for pressing hip seams or wherever there is a curved seam. A sleeve roll performs a similar function and is useful for inserting into sleeves to press seams. A tailor's pressing glove is useful for pressing garments on a dress stand. The padded glove or mitt is worn on the hand and placed inside the garment while the other hand operates the iron. For woollen fabrics a smooth block of wood known as a tailor's clapper can be used to 'block' or flatten an area directly after pressing. The block is held firmly in place for a few seconds after a burst of steam has been directed at the fabric.
Scissors
Scissors vary in size, with 8-, 10- or 12-inch blades; the longer blades are used by tailors and are also known as shears. Dressmaking scissors and shears are available for right- and left-handed users and should only be used for cutting cloth and threads. Three pairs of scissors are needed for the projects in this book: one for cutting out fabric, a separate pair for paper, and small, sharp scissors for cutting threads and buttonholes. For the Edwardian dressmaker the best scissors were those made from Sheffield Steel. An advertisement for the firm Cox & Co. shows cutting scissors with leather covered handles and small buttonhole scissors, which would have been useful for cutting hand-worked buttonholes on delicate fabrics.
Advertisement for a range of Sheffield Steel scissors in _The Lady's World_. (Worthing Museum and Art Gallery)
Seam ripper
No matter how experienced a maker is there is always the possibility of making a mistake when sewing. Edwardian dressmakers used a small sharp knife for ripping seams apart. A razor blade was another option. A tailor I used to work with used a small folding penknife for cutting threads, which hung from a piece of elastic tied to his belt. As an alternative a seam ripper (also known as an unpicker or stitch ripper) is available in two sizes. Small, sharp scissors can also be used.
Tape measure
A good quality tape measure printed in both metric and imperial measurements is an essential piece of equipment. A tape measure should not be wound into a tight swirl in case stretching should occur. Edwardian tape measures were advertised as being 60 inches long. A small sewing gauge is also useful for measuring small areas and for marking hems. The 'Picken dressmaker's gauge' was patented in 1915 and is described in the Woman's Institute of Domestic Arts & Sciences booklet _Essential Stitches and Seams._ It could be used for marking tucks, plaits and ruffles, as well as hems, buttons and buttonholes.
Expandable button spacer
This is a useful tool for working out the spacing of buttons, fastenings or pleats and was not something an Edwardian dressmaker had the benefit of owning. It is lightweight and folds away, and can easily be stored in the bottom of a workbox.
Long ruler
A metre-long ruler is valuable for measuring fabric and is essential for pattern cutting. A folding metre stick is lightweight and convenient for taking into archives (if allowed).
French curve
A flat curved tool used for drawing curved edges. The Pattern Master performs the same role and is another essential tool for costume makers.
Drafting paper
A roll of dot and cross paper marked at 1 inch or 2.5cm intervals is a worthwhile investment. Squared paper marked at 1cm intervals and plain newsprint can also be used.
Hand sewing needles
Hand sewing needles were an essential item of equipment for Edwardian dressmakers, some of whom did not have access to a sewing machine and therefore made whole garments by hand. The size of needle varied to suit each task with short, fine needles used for small stitches. Having the right sized needle was essential to complete each process skilfully and in good time. A good set of assorted needles is useful, with additional needles for specific jobs. For example beading needles are long and fine and fit through the opening of the tiniest bead.
Sewing threads
To replicate Edwardian garments accurately, natural sewing threads should be used. A 1902 advert for sewing thread in _The Drapers' Record_ for Gütermann & Co. shows that the firm produced sewing silk and machine twist. Buttonhole twist was also available – this is a thicker thread used for hand sewing buttonholes. To match the colour of the sewing thread to the fabric, unroll the end of the thread and place it on top of the fabric (this should be done in natural light). If an exact match cannot be found then select a shade darker.
Advertisement in _The Drapers' Record_ for 'REFORM' sewing silks and machine twist for manufacturers. (© EMap and the London College of Fashion Archive)
Fray Check
The Edwardian dressmaker certainly would not have heard of this product but it is invaluable to the contemporary costume maker. A dab of liquid Fray Check can be used to strengthen corners or to stop threads from fraying on delicate work. Other brands are available.
Bodkin
A bodkin is a flat, blunt needle with a large eye, useful for inserting ribbon through lace or elastic into a casing. A safety pin is an alternative suggestion.
Pins
'Cheap pins are not an economy,' warned the Woman's Institute of Domestic Arts & Sciences in _Essential Stitches and Seams,_ and this is still the case. Cheap pins have a tendency to bend, break or rust and therefore stainless steel pins are the best option. Costume workrooms tend to use magnetic pincushions, which are also useful for picking pins up from the floor.
Turning tool
The social campaigner Clementina Black wrote in her study of women's working practices published in 1915, _Married Women's Work,_ of collar makers in London performing the 'turning process' which was the act of turning the collar the right way round once it had been stitched. She observed that it was a tricky process and that most makers used a bone tool to poke the corners out to achieve a neat finish. A bamboo collar turner from Merchant & Mills is a useful modern equivalent.
Beeswax
Sewing thread is pulled through the edge of a block of beeswax to coat the thread with a layer of wax for extra strength. Tailors use beeswax for coating buttonhole twist when sewing a hand-worked buttonhole.
Thimble
A thimble protects the middle finger of either hand depending on whether the maker is right- or left-handed. Indentations in thimbles are there to make it easier to push the blunt end of the needle through fabric. Tailor's thimbles are open-ended because the tailor pushes the blunt end of the needle through layers of cloth and canvas with the side of the finger rather than the tip. The Woman's Institute of Domestic Arts & Sciences advised that the thimble 'should not be tight enough to stop the circulation of blood in the finger, and yet not loose enough to drop off when you are sewing.' The Edwardian dressmaker had the option of purchasing a practical yet decorative thimble from a jeweller or 'fancy dealer'. The Dorcas thimble, for example, was made from three layers of steel and silver, not only to make it more comfortable but also to make it more resistant to damage.
Tailor's chalk and marking pens
Tailor's chalk is available in several colours, in either triangular or squared shapes. The edges of the tailor's chalk should always be kept sharp, either with a tailor's chalk sharpener or by using the blade of paper scissors. An alternative is an air erasable marker although it is advisable to test the marker on a scrap of the fabric before use. Water-soluble markers are also available.
Tracing wheel
There are two kinds of tracing wheel – blunt edged and sharp edged. Both are used to mark seam lines and construction points on fabric; the sharp edged wheel is useful for heavier fabrics. Carbon paper is placed underneath the fabric and all markings can be transferred to the reverse side of the fabric by wheeling along the lines. A self-healing cutting mat, a cork board or a heavy piece of cardboard placed behind will help protect work surfaces.
Light
Electric lighting was a feature used to entice machinists to apply for work in Edwardian factories. Advertisements in _The Drapers' Record_ during the period often stressed this as an attractive feature for workers, alongside adequate ventilation. Sewing black fabric and close beading work still require a good workroom light.
Dress stand
The Edwardian dressmaker was able to buy a variety of dress stands. An advertisement for Clark's dress stands in _The Lady's World_ has a stand with padded, detachable and moveable arms, which sounds very useful. A cage in the shape of a skirt was another optional attachment for keeping the shape of the skirt without the need to add petticoats to the stand. A basic dress stand would be extremely useful for making the projects featured in this book. Stands can be increased in size by padding with wadding.
Advertisement in _The Lady's World_ for dress stands manufactured by John Clark & Co., London. (Worthing Museum and Art Gallery)
Emery bag
A piece of Edwardian sewing equipment no longer in use is the emery bag. The Woman's Institute of Domestic Arts & Sciences included the emery bag in their list of useful workroom tools. It was a strawberry-shaped red denim bag that was filled with emery, a tough material used for polishing metals. The emery bag was used to push a needle through when it became rough or made a squeaking sound.
Haberdashery box
For the wealthier Edwardian dressmakers, smart department stores such as Harrods offered a complete sewing kit in a custom-made walnut box which, like the decorative domestic sewing machine, would have complemented the furniture in the home. The kit contained sewing threads, buttons, needles, pins, mending cards, hooks and eyes, and thimbles. A modern and lightweight alternative is a small plastic toolbox.
Sewing machine
The lock-stitch sewing machine was widely available to the Edwardian dressmaker as either a hand or treadle version, or a powered machine. Machines had basic functions, for example the tension had to be adjusted for each new weight of material. In 1907 _The Ladies' Field_ proclaimed that for the home dressmaker a Singer treadle machine was ideal. In 1908 the company released what was to become its most popular model: the Singer 66k. For those unable to afford the full price of a machine an instalment incentive was available. The weekly hire of a machine was another option. All projects in this book can be made on a domestic sewing machine or an industrial version. A domestic or industrial overlocker is also useful but not essential; the seams of Edwardian garments were either enclosed or overcast by hand.
Edwardian patterns
Edwardian tissue paper patterns had few, if any, pattern markings. Unmarked and unused brown tissue paper patterns remain with a few copies of the sewing journal _Weldon's Home Dressmaker_ at Worthing Museum. The pattern pieces do not contain any balance marks or indication of the grain lines. In a 1901 edition of _Isobel's Dressmaking at Home_ the straight grain is referred to as 'the straight way of the cloth'. The journal also states that directions for making up are stamped on each pattern piece and that if instructions are accurately followed 'success must be insured to the worker.' A small layout diagram to show the position of pattern pieces on fabric was also included. Patterns in this particular journal were only available in a medium size, which is described as 24-inch waist, 36-inch bust and 42-inch hips.
TECHNIQUES
Looking inside garments of the period reveals much about methods of construction. During my research, the discovery of entire garments with pin-tucks and lace insertions entirely made by hand was unexpected; likewise the discovery of many preformed fastenings and trimmings. It seems that dressmakers were either saving themselves time by using readymade goods or spending many hours on small-scale and time-consuming sewing jobs.
Taking patterns from existing garments and accessories
As a result of canvassing the opinion of a selection of experienced costumiers it became clear that experienced cutters and makers take patterns from garments in a variety of ways. With no universal agreement in use I decided to approach each project with the idea of using the method most suited to the garment and to the space I was working in, most usually a table in a museum archive. As the garments used are museum pieces I had to be as careful as possible when handling so as not to cause damage, wearing white cotton or latex gloves. For the accessories – hat, bag, parasol – I began by photographing the item from all angles. I then made sketches and added measurements and notes on details. With the garments, I began by laying each piece flat on a table and photographing the garment with a front and back view. I then photographed details, which included embellishments, fastenings and linings. When possible I mounted each garment on a mannequin for further photographs. I found it useful to sketch each garment full size before sketching details and writing notes. This helped to draw the eye in to details not noticed at first glance. Through trial and error I found it useful to make a pro forma measurement sheet to allow for a methodical recording of details. One striking fact that emerged was that in most cases one visit was inadequate and it was necessary to visit the items a second and even third time.
The patterns were taken using a variety of techniques. One method was to place a garment on a flat surface lined with pattern paper and use fine pins and a pencil to carefully mark the outline of the garment. The exterior details such as frills and flaps were recorded by placing a piece of muslin on top. The muslin was larger than the detail and I pinned around the edge to mark the shape, checking alongside the measurements. I also followed the following advice, given by a range of experienced costumiers: for taking patterns establish the centre front and centre back of bodices and skirts. If possible, use light weights to keep the garment flat. For delicate garments hem weights would be ideal although this method is not advised for museum garments. When the basic pattern has been taken details can be checked (for example ensuring that all seams that meet are the same length). Use a straight edge and a Pattern Master to true up the pattern before cutting a toile. Put garments on mannequins if at all possible and even better use padded arms. It is also a really good idea to write a list of all measurements needed before visiting the museum, to avoid missing any vital information during the visit.
Tacking and basting
In _Cutting Out for Student Teachers,_ author Amy K. Smith explained the process of tacking in 1910:
Tacking stitches must never exceed three-eighths of an inch in length, under this is better. Stitches must be perfectly straight and should be worked very quickly. Commencement and finishing off are best done by back stitches, but if a knot be used it must be small and very neat.
A long thin darning needle was also recommended for tacking, threaded with fine cotton or machine silk in lengths not exceeding 24–28 inches. The problem of thread knotting can be avoided with the following piece of advice (passed on to me when I worked in the tailoring department of an opera house): 'knotting can be avoided by threading the end of the thread which comes from the reel, and not the one broken from the bulk of the cotton, which rubs the fibres of the thread the wrong way, and so encourages knots.' Basting cotton can be used for tacking and the advantage of basting cotton is that it breaks easily. It does not become embedded in the material and can be easily removed with tweezers if necessary. For a stage costume, it is not unknown for troublesome white tacking threads, impossible to remove in black wool fabric, to be coloured in with a black permanent marker.
Tailor's tacking and thread marking
It is essential to have a good solid knot rolled at the end of a length of thread to be used for tacking. To make a tailor's knot twist the thread around the forefinger of the left hand one-and-a-half times. Using the ends of the forefinger and thumb of the right hand roll the thread towards the end of the finger and slip off; the knot is then formed by sliding the knot off the finger using the nail of the middle finger of the right hand.
Tailor's tacks are used to transfer markings between two layers of fabric. Using a double thread this temporary stitch is sewn through all layers of fabric and they look like a person swimming when first made – a round head and two longer arms. The loop is cut and then the fabric pulled gently apart and the inner threads are cut.
Thread marking involves making a long running stitch with a loose stitch on the top. This stitch is then snipped and the fabric is pulled gently apart and the inner threads are cut.
Grain lines
Grain lines were not always rigidly followed by Edwardian dressmakers, economy of material being the most likely reason. One example is the original skirt featured in Chapter 7: it has both back panels of the skirt cut on different grain lines; one is straight at the centre back and the joining panel is cut on the cross. To achieve the best results when making projects in this book it is necessary to follow the grain lines indicated on the patterns. To place a pattern piece on the straight grain of the fabric, fold the fabric in half lengthways making sure both selvedges are lined up. With the pattern piece sitting on top of the fabric place a pin at one end of the grain symbol and pivot the pattern piece until the line measures equal distances to the folded edge of the fabric. The fold is a more accurate straight edge than the selvedge; the selvedge can be uneven due to the manufacturing process.
Seams
Edwardian seams were very narrow. Unlike today, seams often used the selvedge because it was a more economical use of fabric. Open seams were used on the centre back seams of bodices; other bodice seams were narrow open seams pressed together to one side. Narrow flat felled seams and French seams were used on underwear and delicate clothing.
Rolling and whipping
Straight-sided lace was inserted between two pieces of fabric by first rolling a very narrow seam by hand and securing in place with a slipstitch. The lace was then sewn to the outside of the rolled edge by whipping or oversewing. This technique was used on the original Edwardian blouse in Chapter 6.
Hems
Skirt hems were approximately 10cm deep and formed by the addition of a separate, curved band or bias strip sewn to the edge of the hem and folded to the inside of the skirt. An additional braid was sewn to the inside of skirts worn outdoors known as a brush braid or plain wool braid.
Fittings
In a theatre wardrobe ideally two fittings would take place before a costume is finished. The first fitting is a garment in the early stages and the second fitting checks the fit before finishing. The fitting stage is a chance to eliminate any looseness and to check a garment sits in the correct place on the body. The length can also be corrected and any details such as the position of pockets and decorative trimmings can be checked.
Alterations and repairs
To make authentic-looking Edwardian garments evidence of an alteration or a repair could be added. For an Edwardian woman, a small tear or the appearance of an unsightly hole was no reason to discard a much-loved garment. The evening dress featured in Chapter 9 has a hole repaired in the foundation bodice with a patch and the centre back seam has been let out. It is likely that the bodice formed the basis of a previous gown and has been re-used.
DECORATIVE FINISHES
Faggoting
Faggoting was a tiny and narrow decorative stitch used to join two pieces of fabric together. Austrian cotton faggoting is a ready-made alternative.
Illustration of faggoting, a method of joining fabrics together, _Weldon's Ladies' Journal_. (Worthing Museum and Art Gallery)
Bobbles
Dangling, fabric-covered bobbles were a poplar decorative feature on both day and evening dresses. Both the day dress and evening dress featured in this book have decorative bobbles. Bobbles were made by gathering the edges of a circle of fabric over a soft lump of wool. Thread-covered bobbles, firm to the touch, feature along the edge of a belt that is part of a day dress from the collection of Marion May.
Edge of a scarf decorated with a row of thread-covered bobbles. (Marion May)
Embroidery and beading
Beading was a popular decorative feature on Edwardian garments and accessories, especially on net, a lightweight fabric used to counterbalance the weight of the glass and jet beads. Embroidery also featured on garments, accessories and homewares. _Weldon's Practical Needlework_ included embroidery patterns for readers to use at home; one fine example uses a mixture of satin stitch and couched ribbons to form a fuchsia design. Embroidery and beading requires specific tools and materials. Embroidery hoops come in a range of sizes and each suits a different job; for small motifs a small hoop is easier to handle. Small sharp scissors, small needles and a good light are also useful.
Couching
Couching is a way of securing threads to fabric using small stitches to overstich, useful for applying round braids and for sewing rows of beads in place.
Satin stitch
Satin stitch is a raised stitch sewn in parallel lines. It was often used for embroidering names and monograms on underwear, and for embroidering flowers and leaves and other trailing shapes seen on whitework.
Lace making
Women's journals provided lacemaking instructions for readers to make their own lace at home for trimming underwear, children's clothes and edging household linens. _Weldon's Practical Needlework_ featured instructions for making Torchon lace in crochet, which was to be executed in fine cotton with a steel crochet hook.
Tucks and pin-tucks
Tucks and pin-tucks were a decorative feature on many Edwardian garments and underwear. Tucks were also a means for allowing ease in a garment particularly if the tucks fell from the shoulder and stopped before the bust. As social campaigners Clementina Black and Mrs Carl Meyer observed, some factories had machines to sew neatly spaced rows of fine pin-tucks and this process would be done before the blouse pieces were given to the maker. For home dressmakers without this option, pin-tucks were formed by pinching the fabric to create a fold on the straight grain of the fabric and then sewing a running stitch formed of tiny stitches close to the edge.
Dying with tea
White fabrics discolour with age; a garment that looks ivory or cream now may once have been pristine white. If an aged look is the desired outcome white cotton lace or fabric can be dipped in tea (no milk) to achieve an ageing effect. It is advisable to experiment with a sample of tea solution before dipping the actual lace or fabric. It is not advisable to dip larger pieces of fabric because it is difficult to get an even coverage.
TIPS FOR DESIGNING AND MAKING COSTUMES FOR FILM AND THEATRE
This section contains interviews with two highly experienced practitioners, costume designer Frances Tempest and theatrical cutter and tailor Tony Rutherford.
Frances Tempest, costume designer
_How authentic does a costume need to be for film? Is a period 'look' more important than factual accuracy?_
It totally depends on the style of the film, and that's a collective decision created by the whole creative team! Sometimes complete period accuracy is what is required and sometimes an impression of the period is more important. Complete accuracy is out of fashion for costume design at the moment; relating to a contemporary style is the vogue.
_As a designer, what draws you into a character?_
Everything starts with the script. Then wide research, paintings, photos, advertisements, magazines, reading the social history of the period, etc. What one is trying to do is to embed the character into the narrative and tell the story of that character, so ultimately it should be a fusion of historical accuracy and telling details that will inform the audience about the character.
_Finding authentic fabrics, lace and trimmings can be a problem. Would you suggest using vintage fabrics and trimmings if modern equivalents cannot be sourced?_
Using vintage trimming usually helps to give a period look to a garment. Vintage fabrics and trims will behave in a way that modern ones just will not do as they are usually made from natural fabrics. Modern equivalents will sometimes work if they are distressed, boiled, dyed, broken down and generally have the 'newness' knocked out of them.
_How important is underwear to the look of a costume? Would you specify a set of Edwardian underwear, including a corset for an Edwardian dress?_
Silhouette is one of the most important ways in which a costume designer can place a garment into a period. The only way to achieve the correct graphic outline is to have the correct underpinnings. Original garments were made to be worn over a corset and will not work without one! There is also the psychological aspect of the actor feeling more 'in character' if the outfit is correct from the underwear up.
_Do you have any other tips for people interested in designing and making costumes from the Edwardian period?_
I am always struck by the fineness of detail in Edwardian garments and the scale of the trimmings, the fastenings, the embroidery etc. – all finely executed and daintier than anything produced post-1920. It was also the flowering of luxury, revealed in fine fabrics and impracticality: these women relied on others not just to launder and iron their clothes but also to physically dress them. Even the suffragettes, thought to have been practical and modern, appear to have worn what to modern eyes seem ridiculously elaborate blouses.
Tony Rutherford, theatrical cutter and tailor
_How do you prepare wool before use?_
I always steam wool before I start by hovering the iron just above the cloth. Sometimes you can see wool shrink before your eyes! It is an absolutely vital stage of the tailoring process.
_Do you work in imperial or metric measurements?_
I work in both depending on the cutting system that I am using. Older systems like those used in _The Tailor and Cutter_ are in inches and more recent systems like Rundschau use centimetres.
_What is your preferred method for scaling up a pattern from a costume reference book?_
I prefer to use a grid to scale up patterns. I have tried scaling on the computer and printing the pages but a typical Janet Arnold page, for instance, prints out to nearly the size of my workroom wall! I always use the pattern as a guide to the shape and always make a toile to test the pattern before cutting into fabric.
_Which reference books do you enjoy using?_
It really depends on the period I am working on. The two Norah Waugh books are invaluable for any costume maker; _The Cut of Men's Clothes: 1600–1914_ , and _The Cut of Women's Clothes: 1600–1930_. The V&A 'In Detail' books are really useful for close-up details of buttonholes, pockets, pleats and other details.
_If you have to make costumes requiring a quick change what kinds of fastenings do you use?_
Most usually a combination of snap fasteners and Velcro.
_Theatre costumes have to be strong enough to last for repeat performances. How do you ensure costumes will last?_
This really depends on the material but I would usually mount the top fabric onto a backing of cotton Silesia to make the finished garment more durable.
Chapter 3
Fabrics, Measurements and Sizes | 3
---|---
INTRODUCTION
Fabrics
_The fabrics included in this chapter are intended to give just a flavour of what was a vast range of fabric and trimmings choices for Edwardian dressmakers. Suggestions for suitable fabrics for the projects included in this book are made throughout, although in many cases it is not possible to replicate the original fabric, lace or trimming exactly, because they are no longer produced. Buying vintage fabrics and trimmings is an option but they can be time-consuming to source and expensive to buy. The white cotton I used for the underwear projects dates from the 1940s and was recently discovered in a shipping container, where it had been stored many years. It was marked in places but all stains were removed with washing and there are some imperfections in the weave, but this all adds to the authentic feel of the fabric. A list of the suppliers of fabrics, trimmings and equipment used for the projects is included at the end of the book. Fabric widths were not standardized in the Edwardian period and tended to be narrow, so quantities stated in dressmaking journals seem excessive even given that some Edwardian dresses, with full hems and frills, required a lot of material. One example is an evening gown featured in_ Isobel's Dressmaking at Home, _April 1901: 'Ten and a half yards of 22-inch satin...with eight yards of chiffon 44 inches wide'._
A selection of fabrics and trimmings used for projects in the book.
Measurements
Metric units: metres (m), centimetres (cm) and millimetres (mm) are used throughout this book. The easiest way to convert to inches is to use a tape measure marked in both metric and imperial measurements. To convert measurements when drafting patterns, dot and cross pattern paper is recommended because this is marked at 2.5cm intervals, which is equivalent to 1 inch. Pattern paper divided into blocks of 5cm squares is also useful. The patterns included in this book are between a size 10 and 12 and have a 90–92cm bust and a 65–67cm waist. For alternative sizes all patterns can be graded up or down.
FABRICS AND TRIMMINGS
Dressmaking was prevalent in the Edwardian era and fabrics and trimmings were widely available, either from a department store, a local draper or a travelling salesman. The thrifty dressmaker could stock up on fabrics in the twice-yearly sales held in January and July. London department store Marshall and Snelgrove advertised in their 1910 catalogue a sale beginning on 3 January and lasting for three weeks.
Colours and prints
Edwardian fabric advertisements described a range of plain colours with poetic names – myrtle, bronze-green, heliotrope and ruby are just a few examples. Prints and woven fabrics were also popular; the day dress featured in Chapter 8 is made from striped cotton with a jacquard weave. In February 1906 _The Ladies' Field_ reported on the craze for tartans in blouses and underskirts. In 1907 black and white checks of all sizes were popular for 'kilted skirts' worn with long black coats. Gingham and zephyr were manufactured from yarn that had been dyed before weaving and gingham in particular was marketed as a fabric that washed well. Subtle floral prints and sprig prints were also seen, although there is less evidence of their popularity in fashion editorials and advertisements. Royal Pavilion & Museums has a sprig print blouse in shades of blues and greens on a white background in the collection. A floral print dress also appeared in the bound volume of fashion plates produced by Barrance and Ford of Brighton Ltd, in shades of blue, green and pink.
In 1900 _The Lady's World_ described a new pale blue cloth for sale in the Oxford Street department store D.H. Evans & Co. as resembling the blue paper used for telegrams at that time. _The Lady's World_ said the colour was becoming to the complexion, unlike a recent fad for turquoise that was described as having an 'evil effect'. The idea of colours complementing the wearer's complexion was a common theme in advertising and editorials. White was a popular colour for Edwardian blouses; one reason for this was a perception that white was the most flattering colour to have next to the face. This idea appears to have been often enforced by fashion journalist Mrs Eric Pritchard. Writing in _The Ladies' Field_ in November 1906, she claimed:
White, and dead white in preference to cream still holds its own.... I have frequently referred to the fact that dead white is more becoming than cream. Where would the majority of men be without the becoming clean white linen collar and cuffs? It is particularly becoming to the complexion.
As well as being 'becoming' to the complexion, white was also a highly fashionable colour. For example, in February 1908 _The Ladies' Field_ reported that the wearing of white was increasing in popularity among smart and well-dressed women, which suggests that it was a symbol of status for the leisured classes. Black was also a fashionable colour as well as a signifier of mourning.
Floral patterned gown from Barrance and Ford, Brighton, Ltd. (Peter Hinkins)
Relatives of the author wearing mourning dress in c.1905. (Author's collection)
GLOSSARY OF FABRICS
This section includes fabrics used in the book and also a selection of the most popular fabrics mentioned in Edwardian fashion journals and advertising features.
COTTON
There were many different weights and qualities of cotton cloth produced in the Edwardian era, from the lightest weight cottons used for expensive underwear to coarser cottons used for cheaper garments.
CALICO
Plain, undyed and untreated cotton fabric, available in different weights inexpensive to buy and ideal for making toiles.
LONGCLOTH
A lightweight cotton also described as Indian longcloth, suggested in dressmaking journals as an alternative to silk for making delicate underwear.
MUSLIN
A loose weave, lightweight cotton fabric, used for mounting and lining in lightweight garments. Muslin can also be used to make toiles when a lighter fabric is required, for example muslin should be used to toile the blouse project featured in Chapter 6.
NAINSOOK
Nainsook is a fine cotton fabric often used for underwear. It was suggested in Edwardian dressmaking journals, alongside longcloth, as a cheaper alternative to silk when making lingerie or nightgowns.
PIQUE
Cotton or cotton-blend fabric woven with small raised geometric patterns, often used for sports clothing.
SILESIA
Silesia is a tightly woven, strong cotton fabric, which makes it ideal for pockets and for mounting pattern pieces for costumes that need to be robust.
COUTIL
Corset fabric that can be used for a foundation bodice.
DOMETTE
A soft fabric used for interlinings.
FLANNEL
Flannel is an all-wool fabric, popularly used for Edwardian blouses and nightwear. It was often marketed as being 'practically unshrinkable'. An advertisement in _The Drapers' Record_ for Fred Doble & Sons of Dewsbury shows a mother and her daughter wearing long pastel-coloured nightgowns made from flannel in a homely setting.
FLANNELETTE
A brushed cotton fabric in a plain weave, it was soft to the touch and napped on one or both sides. Flannelette was a cheaper version of flannel and was often used for children's clothes.
LACE
Selection of lace used for projects in the book.
Lace was an extremely popular fabric in the Edwardian era. The Edwardian dressmaker could buy instructions for making her own lace. _Weldon's Practical Needlework_ had instructions for making Honiton lace, Torchon and guipure for trimming homemade underwear.
INSERTIONS
Insertions were strips of lace with two straight finished edges, used for trimming blouses and underwear. Insertion laces varied in thickness and width.
GUIPURE
Guipure lace is made when a pattern is embroidered on fabric and the fabric is then removed to leave a thick and heavy lace that stands up in relief. Guipure was either used as a whole piece or the motifs were cut away and applied as trimmings on blouses. Guipure is still available as either fabric or lace trimming.
Postcard of the actress Kitty Gordon wearing a striped day dress with guipure lace motifs applied to the surface, 1904. (Author's collection)
CROCHET LACE
Crochet lace was worked with a small fine steel hook. Many sprigs, usually of leaves or flowers, were worked over a foundation cord. The spaces between the sprigs were joined by crochet bars. Irish crochet was particularly singled out for being of exceptional quality in fashion editorials.
VALENCIENNES
This is a flat bobbin lace worked with one thread which forms both the design and the background.
CLEANING LACE BLOUSES
The _Every Woman's Encyclopaedia_ provided tips on cleaning delicate lace blouses, which could not be laundered with soap and water. They recommended sousing in petrol, with the heavily soiled parts such as the collars and cuffs being scrubbed with a soft brush. It was then suggested that garments were squeezed rather than being wrung before being carefully pulled back into shape. The smell of petrol was said to disappear if the garments were left to dry in the open air for at least an hour. To complete the process the delicate lace garments were to be ironed and then stuffed with tissue paper before being stored away. Readers were further advised that holes could be mended by darning and that frayed collars and cuffs should receive new bindings. Petrol in the Edwardian era was not the same consistency as the fuel we are familiar with now.
TULLE AND NET
Net was used in Edwardian garments and accessories as a base for beading and for creating braids, which can be seen on the evening gown in Chapter 9 and the velvet toque in Chapter 11. Tulle is a finer form of net and was available in silk. Machine-made figured tulles such as _point d'esprit_ were a popular choice for blouses towards the latter part of the Edwardian period. Writing in _The Lady's World_ in 1900, 'Madame Modish' extolled the virtues of tulle for trimming a hat: 'Tulle is very perishable, but at the same time easily manoeuvred by the amateur'. She suggested combining colours to create a shading effect, for example pinks and mauves.
LINEN
Lightweight linen was used for blouses and summer dresses, while coarser weight linens were used as interlinings.
NUN'S VEILING
This is described in _Cutting Out for Student Teachers_ as 'A thin, rather loosely woven plain material, with somewhat a rough dry appearance although soft to the touch. It washes admirably, and is durable considering its texture.' Collector Marion May has a nightgown with a square neck from the late Edwardian period with a label showing that it is made from nun's veiling. It feels soft to touch and would have kept the wearer warm through the night.
Nightgown made from nun's veiling with a square neck embroidered with whitework. (Marion May)
SATIN
_The Ladies' Field_ advised readers in October 1907 that satin was the ideal fabric for a 'house-frock', which if worn with an embroidered scarf and underbodice of lace or chiffon could even be suitable as a tea-gown to wear in the late afternoon.
SERGE
Serge was a hardwearing fabric used for outdoor clothing. An advert from Fred Doble & Sons of Dewsbury featured in _The Drapers' Record,_ 31 May 1902, for their registered fabric 'storm serge' promised, 'sea water will not affect the colour'.
SILK
A variety of silk fabrics was available to the Edwardian dressmaker. Foulard was a soft twilled fabric used for dresses. Silk moirette was said to make a good petticoat. Lightweight silks were imported from China and Japan and used to make blouses. Widths of fabric varied between 21 inches and 36 inches. Chiffon was an opaque fabric originally made from silk but is now available made from artificial fibres.
VELVET
Both silk and cotton velvets were used for Edwardian garments and accessories. Velveteen was also advertised. In 1909 _The Lady's World_ cautioned against using a heavyweight velvet for hat making and suggested 'millinery weight' must be purchased. Unfortunately it is no longer possible to buy millinery weight velvet. Velveteen was a less expensive alternative to silk velvet.
Advertisement for 'Lewis's Wonderful Velveteen', a cheaper alternative to silk velvet, available to buy for two shillings a yard from Lewis's in Manchester. _Weldon's Home Dressmaker_ , No. 112. (Worthing Museum and Art Gallery)
VIYELLA
A trade name for a lightweight plain weave fabric composed of 45% cotton and 55% wool. Viyella was marketed as being pre-shrunk and therefore washable.
WOOL
Many varieties of wool were on sale to the Edwardian dressmaker; one lightweight example is Delaine, which was popular for winter blouses. Tweeds were popular for winter-weight tailor-made costumes and cloaks. The subtle colours of tweed fabric were much admired by _The Lady's World_ in 1909; they could find not a single unpleasant shade.
INTERFACINGS AND BONING
From handling many Edwardian garments in museum collections it seems Edwardian dressmakers used little in the way of interfacing in lightweight clothing. Plackets, cuffs and facings are soft to the touch rather than firm, which would be expected if interfacing had been used. In jackets and costume bodices a layer of canvas was used to structure collars, cuffs and facings. Boning was often used, however, and can be found in collars, at the centre back of waistbands and attached in casings along the seams of bodices. Boning was of course used for shaping corsets, although some corsets designed specifically for playing sport were unboned.
PETERSHAM
When a straight waistband was applied to the waist of a skirt the method of adding structure was to insert Petersham ribbon. The original skirt featured in Chapter 7 has a waistband stiffened with Petersham ribbon.
BRUSH BRAID
Brush braid was designed to protect the hem of an outdoor skirt and was hand sewn just inside the hem of the skirt all the way around the hem. An advertisement in _The Drapers' Record_ in April 1902 shows braid made from either mohair or silk that was 'splendid wearing'.
Advertisement in _The Drapers' Record_ for 'elegant' skirt facings with silk plush cord and real mohair brush edge. (© EMap and the London College of Fashion Archive)
Advertisement for Cash's Cambric Frillings, _The Queen_ , 1907. (Worthing Museum and Art Gallery)
Trimmings
In her book _Shops and Shopping, 1800–1914: Where, and in What Manner the Well-Dressed Englishwoman Bought Her Clothes,_ Alison Adburgham explained that the haberdashery counter in a store was always located on the ground floor near to the entrance. It was understood that a customer might spend a small amount of money on trimmings and might then venture into the rest of the store and make larger purchases. The smart London department store Harrods had a Lace and Ribbon Department that was situated on the ground floor. According to a book commemorating the firm's diamond jubilee in 1909, customers could find
... rich examples of laces from all parts of Europe...Ribbons, Fancy Neckwear, Feather Stoles [and] departments devoted to haberdashery, trimming, Fancy Needlework, Artificial flowers, Parisian jewellery, Ladies' Hosiery, Gloves and shawls, the whole forming a galaxy of beautiful colour and fabric of extreme charm.
Edwardian clothing is known for its elaborate trimmings, frills and flounces, which gives the impression that dressmakers spent hours frilling lace, adding embroidery and layers of decoration. Some may have taken this route but there was an alternative – readymade frilled insertions. Cash's Coventry Frilling was advocated by dressmaking teacher and author Amy K. Smith in 1910:
It has a gathered thread at the edge which enables the material to be drawn up, so as to make a regular distribution of fullness. The frilling can be bought ruffled into a straight featherstitched band, which is a great saving of time, as the frilling is just ready to be sewn or machined to any garment.
It was reasonably priced, versatile and good quality and, crucially for undergarments, it could be washed and ironed. An advertisement for Cash's 'cambric frillings with drawing cord' features in _The Queen_ in 1907 and readers were advised that it was of the finest quality, was durable and would last a long time. In 1908 _The Ladies' Field_ extolled the virtues of Cash's colour-fast 'washing ribbons' available in two widths in shades of red, pink, blue, heliotrope and white.
Buttons and other fastenings
Decorative buttons were seen on women's clothing but were not used as often as hooks and bars or hooks and eyes as garment closures. The original walking dress featured in Chapter 7 has small flower-shaped jet buttons.
TAKING MEASUREMENTS
Edwardian women wore layers of underwear, including a fitted corset. Measurements taken for most of the projects in this book should be taken over appropriate foundation garments, but the drawers and chemise measurements are to be taken directly from the body. Use the following advice for taking measurements, record all measurements and include the name of the wearer.
**Height** – stand against a wall and with a ruler on top of your head make a small pencil mark on the wall. Measure from here down to the floor.
**Collar** – measure around the base of the neck (not tightly).
**Bust** – measure around the fullest part of the bust. The tape measure must be straight across the back.
**Waist** – tie tape around natural waistline (bend to the side to find this). Measure the waist firmly over the tape.
**Hips** – measure 7–9 inches or 18–23cm down from the waist tape at the side of your body to find the fullest part of your hips. Firmly measure around hips, keeping the tape the same distance from the waist all the way around.
**Nape of neck to waist** – tip the head forward to find the most prominent bone at the top of the spine and measure from here down the spine to the back waist.
**Across back** – measure across the back, passing the tape measure horizontally over the peak of the shoulder blades.
**Shoulder** – measure from the bone at the end of the shoulder to the base of the neck.
**Sleeve** – there are two methods for doing this: either measure from under the arm to the wrist then take off 3cm, or slightly bend the arm and measure from the tip of the shoulder to the wrist bone.
**Bicep, elbow and wrist** – measure quite loosely.
**Waist to floor** – measure from the side front waist to the floor. The finished hem measurement can also be taken in this way.
**Head** – measure around the circumference of the head.
Front view illustration showing where to position the tape measure when measuring the head, bust, waist, hip, waist to floor, bicep, elbow and wrist.
Back view illustration showing where to position the tape measure when measuring the collar, across back and nape of neck to waist.
Sizes
By observing museum garments, clothing advertisements and dressmaking patterns it can be noted that Edwardian women came in a variety of shapes and sizes. For example, a skirt advertisement from 1909 lists available hip sizes as being 36–42 inches. An online clip from 1917, briefly showing a fashion parade that included a gown by the British couturier Lucile, shows tall models, known as mannequins, of an above average size. The Brighton department store Leeson and Vokins offered a made-to-measure service for customers of all sizes; their garments were advertised as being 'made in several sizes...to fit any figure, including extralarge sizes'. In 1895 _Home Chat_ listed their paper patterns as being available to order in three sizes; the largest was bust 40 inches and waist 28 inches, and the smallest was bust 34 inches and waist 22 inches. The length of skirt was a standard 41 inches for all sizes.
Chapter 4
Split Drawers and Chemise | 4
---|---
_Due to Edwardian conventions of etiquette and modesty it is difficult to ascertain just how many women preferred closed drawers and how many still wore split drawers. Advertising features had a tendency to show the drawers folded into a neat square with just the frilled edge around the bottom of the legs on display. Fashion articles in women's journals also tended to sidestep the issue and therefore the dilemma for this chapter was whether to include closed or split drawers._ Weldon's Ladies' Journal, _July 1905, features a pattern for a pair of 'closed knickers', which it acknowledges are preferable to many women, with suggested fabrics being linen, alpaca, silk and thin serge. If a nonwashable fabric was used, women were advised to make a removable lining from cambric or nainsook. The knickers were said to be useful for cycling and boating as well as general wear._
In the hope of resolving the issue I made an appointment at The Clothworkers' Centre for the Study and Conservation of Textiles and Fashion at the Victoria and Albert Museum in London to view a selection of Edwardian underwear. Three pairs of drawers that I viewed in the archive were French-made split drawers with wide, knee-length legs. Each had a deep flounce around the hem with either lace edging and/or lace insertions. One pair, made from fine cotton lawn, had a narrow casing with 32mm wide eau-de-nil satin ribbon threaded through and tied in a bow that sits at the outer edge of the leg. A British-made pair of drawers, dated 1909 and made from wool, was also open between the legs, and so a decision was made to include split drawers in this chapter. While looking at the fabulous collection of underwear in the archive another dilemma arose, which was the subject of which order the underwear was placed on the body. One set of lingerie had a frill around the hem of the chemise making it unlikely that it would be tucked into a pair of drawers. Fortunately _The Woman's Institute of Domestic Arts & Sciences_ lingerie-making booklet provided an answer and suggests that the preference depended on the nationality of the wearer. In France the chemise was worn next to the body whereas American women preferred to wear a knitted vest next to the body. The French chemise was a closer fit and a little shorter while the American chemise was looser and longer and worn over the corset.
Reproduction split drawers and chemise.
SPLIT DRAWERS
The split drawers copied in this chapter are the only project not to be based on a museum garment and were instead bought in an online auction as part of a job lot of assorted vintage clothing. The date is unknown, although they are in keeping with drawers from the late Victorian and early Edwardian period. The owner of the drawers is identified in ink on the outside waistband as G.J. Thomas and a laundry number (c443) appears in faint ink on the inside of the waistband. The edging at the hem is a particularly fine example of machine-embroidered whitework, or white embroidery on white fabric. The scalloped edge of the frill features a repeating floral design while the band of insertion features a symmetrical wave with small flowers either side. The drawers are tied with long cotton tapes at the centre back. Both museums have several pairs of drawers but it was difficult to match up the lace or white-work and so this pair has been included as the only non-museum item. Making a pair of drawers was one of the topics covered in _Cutting Out for Student Teachers_ written by Amy K. Smith in 1910. The amount of fabric suggested for a pair of drawers was given as 1–2 yards. To make the drawers to the right size readers were advised to take a waist measurement with added ease; this was to be taken without a corset being worn. Comfort was the main concern when making drawers, with plenty of room being left at the knee, as well as the drawers having a wide and deep seat. It was advised that fullness at the waist should be distributed towards the back and a shaped waistband was suggested as being more comfortable than a straight band.
Illustration showing the front view of a pair of split drawers from the author's collection.
Knee band and frill detail from the original pair of split drawers.
Split drawers pattern.
MATERIALS AND EQUIPMENT
2 metres medium weight white cotton fabric
Thread
1 metre broderie anglaise insertion lace for the leg bands (optional)
110cm × 1.5cm cotton tape for the waist ties
1.5 metres narrow cotton tape for finishing the inside of the knee bands
120cm scalloped broderie anglaise to edge the legs
Sewing machine
Layout and cutting
Fold the fabric in half lengthways, making sure both selvedges are lined up. Before pinning the pattern in place, make sure it is lying on the straight grain of the fabric. To align the grain line with the straight grain of the fabric, place a pin at one end of the grain symbol and pivot the pattern piece until the line measures equal distances to the folded edge of the fabric. Cut out two legs and two waistband sections. The original drawers have a knee band of white-work or broderie anglaise insertion. I was unable to source a copy of this insertion and so I have cut a band from the white cotton fabric. Add seam allowance to all edges (I added 2cm to all seams and 1cm to the bottom of the leg and to the top of the leg bands). Identify the front of each leg piece; a note with the word 'front' pinned to each section can be useful, especially when attaching the leg bands. It is also worth identifying the centre front and centre back and the top of the legs.
Making order
LEG BANDS
The legs are seamed in one process once the leg band and frill have been sewn to the lower edge. Begin by preparing the frill. Machine two rows of gathering along the top of the broderie anglaise leg trim using a long machine stitch; sew within the seam allowance if possible otherwise the gathering will have to be removed once the frill is attached. To check the length of gathering on the frill, and to distribute the gathers evenly, place a straight rule or flat tape measure on the ironing board and spread the gathers out next to it with the raw edge running alongside the measure, and pin to the board at either end. Once the correct length has been established the loose threads at either end can be secured either by knotting or by wrapping the long threads in a figure of eight around a vertically placed pin at either end. To join the gathered section to the leg band, with right sides facing, line up the raw edges and machine the trim to the band. Trim and press the seam upwards into the band. Gather the bottom of the legs to fit the knee bands. With right sides facing, line up the raw edges and machine the band to the leg and press the seam downwards into the band. Trim the seams to 0.3cm and press lightly, trying not to flatten the gathers. Narrow cotton tape is sewn on top to enclose the seams. Pin the narrow cotton tape on the reverse side to cover all raw edges and topstitch close to the edges along either side. As an alternative a facing made from the same band pattern piece could be hand sewn to the back to enclose the raw edges once the legs are made up, although this adds an extra layer of bulk.
LEGS
The next stage is to make up the legs as two separate cylinders. The edges of the split part of the drawers are finished before joining the lower seams. Fold the edge of the split part of the drawers over by 1cm, press and fold again by a further 1cm and press. Pin in place and topstitch close to the folded edge. The machines used to sew the original drawers would not have had an adjustable stitch length and so the stitches are tiny. A small machine stitch should be used to match the reproduction drawers as closely as possible to the original pair.
Flat felled seams are used to join the inner leg seams on the lower part of the drawers. To make a flat felled seam, with right sides facing line up the raw edges and pin and machine along the seam line. Press the seam allowances to one side. Trim the lower seam down to 0.5cm and the upper seam to 1cm; the seams on the original garment are slightly narrower at 4mm. To complete the flat felled seam, fold the upper seam allowance over the lower trimmed edge and press in place, pin and tack through the centre. Slowly topstitch the seam in place by machining through all layers, sewing close to the folded edge.
Fold the edge of the upper part of the leg over twice and pin in place.
Flat felled seam on the inside leg.
WAISTBAND
Join the two waistband pieces together down the centre front seams. Trim and press the seam open. Finish the two shorter sides by turning twice and topstitching in place with a narrow hem. Using a long machine stitch, sew two rows of gathering within the seam allowance along the top edge of the drawers, leaving a gap at either edge. The drawers are overlapped at the centre front and secured together before the waistband is attached. With right sides facing place one section on top of the other and overlap by 1cm or the width of your seam allowance, machine across the top and within the seam allowance, to hold in place.
With right sides facing and with the raw edges lined up, working from the right side of the drawers place the top to the front lower edge of the waistband. Pull the gathering stiches until the legs match up with the waistband. (This can also be done by pinning the outer edges of both sections flat to the ironing board.) Distribute the gathers evenly, pin in place and machine along the seam line; press the seam into the waistband. Before the waistband is sewn down a casing needs to be formed for the cotton tapes at the centre back. Press the waistband in half lengthways to mark the top edge of the drawers and then open flat. With the tape running along this line position each tape in 10cm from the finished edge and tack in place. Fold the waistband over and topstitch 9cm in from the edge to hold the tapes in place. Machine a further line of stitching under the tape and 2cm down from the folded edge. To finish the drawers, turn the raw edge of the waistband under so that it sits a fraction below the seam line, and pin and tack in place. Working from the right side of the drawers topstitch close to the bottom edge of the waistband.
Detail showing the centre back casings of the drawers.
CHEMISE
In the Woman's Institute of Domestic Arts & Sciences _Underwear and Lingerie_ booklet the chemise is described as a favourite undergarment of many women. The chemise was a roomy garment designed to be tucked into a pair of equally roomy drawers. From observing many variations of the chemise in museum collections I have noticed that the front and back of the garments tend to be the same size, although occasionally the front neck is slightly lower than the back. The chemise was cut on the straight grain and was shaped at the sides. Short sleeves, if added, were loosely fitted. The front bib area and corresponding back area were gathered or tucked into a band at the neck or yoke. A chemise is finished at the neck, most usually with a straight band and a gathered frill. Some have a placket opening at the front and this also may have a frill inserted around the edge. In _Cutting Out for Student Teachers_ Amy K. Smith extols the labour-saving qualities of ready-made trimmings: 'Messrs. J. & J. Cash, of Coventry, prepare various styles of their frilling ready gauged to a feather-stitched band for trimming underwear. This frilling is an immense saving of time.' The shoulders were often placed on a fold and so were straight rather than sloped. Hems on the examples observed were also straight rather than curved. Looking at patterns provided on grids in both _Cutting Out for Student Teachers_ (1910) by Amy K. Smith and _How to Make up Garments_ (1907) written by Agnes Walker, it can be noted that the reason for the straight shoulders and hems was economy of material – a curved hem would mean wasted fabric. A straight hem would also have been easier for a less experienced dressmaker to make up. The chemise was one of the first things a child learned to make at school. Royal Pavilion & Museums, Brighton & Hove, have a beautifully made, reduced scale chemise made by an eleven-year-old child. The garment was exhibited in the International Exhibition of 1871 in a section entitled 'Specimens of school-work'. The dressmaking educator Agnes Walker provided instructions for readers to make a chemise by proportion for women and children. For example, to find out the length a measurement should be taken from the top of the shoulder to the length required below the knee.
Illustration of the front view of the chemise from the collection at Royal Pavilion & Museums showing the shape of the tucks and three-dot broderie anglaise trim.
MATERIALS AND EQUIPMENT
2.10 metres medium weight white cotton fabric
1.70 metres scalloped broderie anglaise to trim the neck and arms
Thread
Ruler or Pattern Master
Chemise pattern.
The chemise chosen for this project is a white linen chemise from Royal Pavilion & Museums, Brighton & Hove, museum accession number C001011. Made entirely by hand the chemise has slight sleeves trimmed with straight broderie anglaise and a wide round neck finished with a band that is trimmed with straight broderie anglaise. It has sixteen vertical bands of tucking in the front bib area while at the back the fullness is gathered into the neck band. The name 'E. Brown' is handwritten in ink at the centre back neck, which would have identified the garment's owner during the laundering process. Although the museum's chemise is complete and is without repairs, it is stained in places and in need of conservation.
Cutting
This basic chemise pattern can be adapted to fit a range of sizes by varying the number of tucks inserted at the centre front and by dropping the armholes. The back is gathered into a band and so can be adapted to fit the tucks at the front. The tucks are not marked on the pattern but instructions are given for spacing and inserting them. To follow the cutting instructions featured in _Cutting Out for Student Teachers,_ fold the fabric in half lengthways and then in half widthways. Place the pattern with the shoulders, centre front and centre back placed against folds. If preferred the fabric can be folded in half lengthways only and an additional seam allowance added to the shoulders. If this method is used then the shoulder seams are joined together with a flat felled seam once the tucks have been inserted in the front and back. Add 1.5cm to all seams; the hem is folded up by 2.5cm with a further 1cm turning and this is included in the pattern, although a longer or shorter length may be preferred. Mark the centre front line with a line of tacking to indicate the desired length of the tucks. Mark the centre back also.
Making order
Begin by making the tucks on the front bib; because the centre front of the bib has a row of tacking already marked at the cutting stage, the edges of each tuck can be measured to radiate out from this line. The number of tucks inserted depends on the size of the wearer. For a size 10–12 chemise insert a total of eighteen tucks, nine either side of the centre front line. These lines form the fold of each tuck. Mark them either with a line of tacking or with an air erasable marker. Using a ruler or a Pattern Master measure straight vertical lines out 2cm both ways from the centre front. Continue to mark at 2cm intervals until a total of eighteen tucks have been marked. The centre four tucks measure 16cm in length and the other tucks decrease in size by 1cm so that the final tuck measures 11cm. To mark the tucks press down the centre of each line. Starting from the centre front sew one tuck at a time by machining 0.5cm in from the fold. Press each tuck out of the way before sewing the next one. When all tucks have been sewn press them towards the centre front to match the Royal Pavilion & Museums chemise, or press away from the centre front if preferred.
Plot the tucks outwards from the centre front.
BACK GATHERS
When the tucks have been inserted measure across the top of the tucked section – this is the measurement to be used for gathering the back neck. To gather along the back neck, set the machine to the longest stitch and sew two parallel lines approximately 0.3cm and 0.6cm apart, close to the seam line. Knot the four threads at one end and pull to gather using the top threads at the other end. Check that the measurement matches the width of the tucks and knot the threads at the other end. Distribute the gathers evenly and using only the tip of the iron, press along the gathers, within the seam allowance.
To make the neck band first join the short edge with a 1cm seam and press the seam open. The seam of the neckband is positioned at the shoulder rather than the centre front or centre back. Pin the neck band to the chemise with right sides facing. Starting at one shoulder line up the raw edges of the band with the top of the neck and, working from the chemise side, pin along the neck and machine along the seam line. Press all seams upwards and trim.
Add the broderie anglaise trim to the neckband by placing right sides together, lining up the raw edges and machining within the seam line. Press the seams downwards. Take the facing band and press the long edges over by 1cm. Join at the short edge with an open seam and place behind the neck band; pin and tack in place and working from the right side topstitch the facing in place.
Pull the top two gathering threads at the centre back.
To add the broderie anglaise trim to the sleeves place right sides together, lining up the raw edges and machine within the seam line. The join is neatened by a flat felled seam. To create a flat felled seam, with right sides facing line up the raw edges and pin and machine along the seam line; press the seam allowances to one side. Trim the lower seam down to 0.5cm and the upper seam to 1cm. To complete the flat felled seam, fold the upper seam allowance over the lower trimmed edge and press in place, pin and tack through the centre. Slowly topstitch the seam in place by machining through all layers, sewing close to the folded edge.
To finish press the hem over by 1cm and up again by a further 2.5cm. Pin and topstitch the hem by machining close to the folded edge.
ADAPTING THE DRAWERS AND CHEMISE
Royal Pavilion & Museums, Brighton & Hove, have a pair of split drawers that have been altered to form closed drawers. The original stitching has been unpicked but the indented holes from the previous machine stitches are still visible. The new drawers have elastic threaded through the waist. For a quick change in the theatre Velcro could be sewn to the back waist opening of the drawers. The chemise is a really versatile garment that can be adapted in a number of ways. The better quality garments viewed during research were made from the finest, lightweight cottons and had layers of insertion, embroidered names and monograms and narrow ribbons threaded through eyelet ribbon around the neck. The lesser quality garments are made from stiffer cottons – one example could have stood up by itself. The chemise could be altered by adding a placket at the centre front and a single button at the neck edge. The number of tucks could be varied, and the trim. For a quick change in a theatre it would help to cut the armholes looser.
Add the neck band.
Neaten the edge of the sleeve with a flat felled seam.
Chapter 5
Flounced Petticoat | 5
---|---
_Sound advice on the length of a petticoat was offered by fashion journalist 'Emele', writing in_ The Lady's World _in 1909:_
It is always a mistake to have petticoats long, they so soon get soiled and draggled-looking, if they are. Just to the ankles is the length for a petticoat, not an inch longer – unless, of course, you are the unfortunate possessor of very ugly feet.
The Woman's Institute of Domestic Arts & Sciences booklet _The Dressmaker and Tailor's Shop_ advised that it was the duty of a dressmaker to guide the customer to choose the right petticoat: 'Frequently the customer does not realize the importance of securing the right kind of petticoat to wear with a beautiful dress or suit, and it should be the duty of the dressmaker to give her some advice about this matter.' When it came to cutting a petticoat it was suggested that the pattern used for cutting the outside skirt could be used for a petticoat, although the fabric must be suitable. The dressmaker 'should always caution her customer against the wearing of a petticoat that clashes with the outside material or that is too scant or too full for the dress itself, for the petticoat can very easily spoil the effect of an outer garment.'
Reproduction petticoat.
Advertisement for a value silk petticoat, the 'Marvel', by Peter Robinson's, Oxford Street, London. Illustrated by Ida Pritchard. (Worthing Museum and Art Gallery)
For evening wear, frilled petticoats were made from 'rustling silk' – this added an auditory element to an evening ensemble. Worthing Museum have a glacé silk petticoat dated 1900–1908; this gored full-length petticoat is fitted at the waist and flares out at the hem. A decorative trim with a serrated edge, also made from glacé silk, has been pleated and sewn to the skirt in a wave around the hem. Taking the petticoat out of its storage box and wrapping of acid-free tissue paper it is striking how much the crunch and rustle of the silk gives an indication of what it must have been like to wear. Frills already constructed with either tucks, pin-tucks, lace insertions or faggoting could be purchased from haberdashers. In 1902, the weekly trade publication _The Drapers' Record_ featured an advertisement listing twelve types of ready-made frilling for sewing onto underwear, which was said to be a way of saving both time and money, and also of avoiding the 'vexatious' activity of making frills. The demise of the full frilled petticoat came with the rise in popularity of the new Empire-line silhouette, although the petticoat was still decorative and considered a vital garment. In 1912 Peter Robinson's in the West End of London were advertising a slim-line petticoat called 'The Marvel', which was sketched for the advertisement by fashion illustrator Ida Pritchard. The peach-coloured petticoat was made from silk taffeta and was finished at the hem with a deep flounce of box pleats. It was also available in a wide range of colours including ivory, turquoise, pink, pale blue, grey, light brown, dark brown, red, navy, light navy, black, and white.
Illustration of the original petticoat from the collection at Royal Pavilion & Museums, Brighton & Hove.
Eau-de-nil satin ribbon threaded through wide eyelet lace around the skirt of a satin and tulle petticoat. (Worthing Museum and Art Gallery)
The petticoat featured in this chapter is from the collection at Royal Pavilion & Museums, Brighton & Hove, museum accession number H56/21/69. It is completely sewn by hand with tiny stitches, a fact that only became obvious when photographs taken in the archive were enlarged on the computer. Although the petticoat is not dated it resembles petticoats featured in advertisements from 1906 to 1908. The petticoat is formed of a main full-length skirt with an additional flounce sewn on top halfway down. The main petticoat is made from white cotton and it is shaped to fit the body with five gored panels. The centre front panel drops in a slight curve at the waist; it has a vertical buttonhole at the centre front, which was possibly used to fasten the petticoat to another garment. The name 'Adrienne' is embroidered in delicate satin stitch to one side of the buttonhole. The petticoat is flat at the front and smooth over the hips, a small amount of fullness is gathered at the centre back in small pleats and it can be pulled in and gathered further when tied. Ties made from cotton tape are threaded through two small channels at the centre back. The ties are long enough to tie once at the back and bring around to the centre front to tie in a bow. Three different laundry markings are crudely sewn in red and blue thread at the back waist.
Detail showing a band of weighted cotton tape sewn behind the eyelet lace of the original petticoat.
The outer flounced layer of the petticoat hangs off a straight piece of crocheted eyelet lace that would originally have had a ribbon threaded through and tied in a bow at one side – this is now missing. The ribbon would have been removed for laundering and kept separately and perhaps it was never replaced after its last wash. Petticoats from a similar date in Worthing Museum and The Clothworkers' Centre for the Study and Conservation of Textiles and Fashion have eau-de-nil or pale blue ribbon threaded though eyelet lace. A piece of cotton tape with round, flat, fabric-covered weights attached at intervals runs behind the eyelet lace. This would have prevented the lightweight and very full petticoat from riding up.
Detail showing layers of lace in the outer flounce of the original petticoat.
The layer of lightweight cotton fabric sewn to the bottom of the eyelet lace is comprised of rows of vertical pin-tucks. There are three lace insertions below this, each 4.5cm wide, which are inserted in stripes around the outer layer. A useful guide to identifying the lace used on petticoats is lace expert Pat Earnshaw's book _A Dictionary of Lace._ The lace on the museum's petticoat appears to be 'Valenciennes', which by the Victorian period was being produced in France and Belgium and is described as 'a neat durable lace used for undergarments' favoured by European aristocracy. Valenciennes can be recognized by its diamond-shaped mesh and curving patterns of flowers and fronds. Valenciennes lace was a popular trim for undergarments and it was often mentioned in society journal _The Ladies' Field_ and the French fashion journal _La Mode Illustrée._ Having established that the petticoat was beautifully crafted and made from expensive lace it perhaps comes as no surprise to find a darned section at the back of the hem – perhaps this is where the heel of a shoe caught the edge of the flounce. With so much care and attention going into the making of the petticoat it was obviously not going to be discarded just for the sake of a small hole.
MATERIALS AND EQUIPMENT
4 metres white medium weight cotton for the main petticoat
4 metres white lightweight cotton for the flounce
12 × 4.5cm metres insertion lace
Eyelet lace and satin ribbon (optional)
1.80 metres crocheted eyelet lace or straight cotton lace, 1.5–3cm wide
3 metres × 1cm wide picot edging lace
1.60 metres × 12mm white cotton tape
Thread
Cutting
PETTICOAT AND FRILL
The petticoat should be cut out by following the grain lines on the pattern. 1.5cm seam allowance should be added to the top and sides with 6cm added to the hem. To face the waistband a bias strip measuring 80cm × 5cm is needed, which includes seam allowance. The lower frill is cut along the selvedge and measures 8cm × 3 metres, which also includes seam allowance. Thread mark the position of the flounce with tailor's tacks.
SEPARATE FLOUNCE
The top flounce is 4 metres × 36cm when finished, with a lower frill, which is sewn to the edge of the flounce, measuring 4.5 metres × 10cm when finished. To make the flounce, join widths of fabric together with tiny French seams to make a 4-metre length, or cut a straight piece along the selvedge. A less full flounce can be made if the maker would rather not sew so many pin-tucks in the upper layer. Three strips of insertion lace measuring 4 metres each, and one piece of crocheted eyelet lace, or straight cotton lace are also needed.
Petticoat pattern.
Making order
Begin by making the darts in the front panels. Working on the wrong side of the fabric fold the dart down the centre, pin in place and machine from the wide end to the point. Each dart is 7cm long and is pressed towards the centre front when completed. The next stage is to join the seams. On the original petticoat a tiny 3mm French seam is used. To make a French seam place wrong sides together and machine so that the raw edges show on the right side of the petticoat. Trim the seam allowance away close to the line of stitching and, to replicate the original seam, trim the seam down to 2mm. Press the seam to one side and then fold along the seam from the right side with the raw edges enclosed inside. Pin and machine a 3mm seam. Press the French seams away from the centre front. The centre back also has a French seam up to the opening. The centre back opening can be finished off by turning the fabric over twice and topstitching in place.
Form the pleats at the centre back following the directional arrows on the pattern. To finish the raw edges at the top of the petticoat, a facing cut on the bias is sewn along the inside of the waist. A space is left open in the back panels only to form a casing, which a cotton tape passes through in order to gather the waist in to fit the wearer. To make the facing, fold the bias strip in half lengthways and press; measure against the top of the petticoat to check the length and machine down the inside of both short edges. Turn the right way round and press. Place the strip along the outside of the waist with the raw edges lining up. Machine along the seam line and trim and layer the seam to eliminate as much bulk as possible. Fold the facing to the inside of the petticoat leaving a 2mm line of facing showing on the right side. Pin in place and sink stitch to form a line of narrow piping. Before machining the bottom edge of the facing, pin the cotton tapes to the side back seams and then machine to the seam allowance. Keep the tapes out of the way and pin and machine along the bottom of the facing, sewing close to the fold. To finish the hem of the petticoat press the hem over by 1cm and up again by a further 5cm. Pin and topstitch the hem by machining close to the folded edge.
The position of the pleats at the centre back of the petticoat.
The position of the cotton ties at the back waist, which are sewn to the seam and then encased in the facing.
Back detail of the finished petticoat.
Adding a frill to the lower edge of the main petticoat.
FRILL
Frills are applied to Edwardian underwear using two methods: the first is to place two rolled edges together and to overstitch by hand using small stitches that at times are barely visible to the eye. The second method is to sew a frill upside down and then press it downwards. This is the method used to sew the bottom frill to the reproduction petticoat.
To make the frill, machine the straight edge of the picot lace on top of the selvedge. Join the ends of the frill with a French seam and press to one side. Mark the frill into four sections and using the longest machine stitch sew a double row of gathering in each of the four sections. Mark the hem into four corresponding sections. Pull the gathering threads until the frill fits the petticoat and secure the ends of the gathers. Press the gathers along the raw edge, pressing within the seam allowance and using the tip of the iron only. Place the frill upside down around the outside of the hem, with right sides facing and the raw edge of the frill butting up against the row of machining holding the hem in place. Pin the frill in place, distributing the gathers evenly as you go. Machine the frill to the petticoat, trim all threads and smooth the frill downwards and press lightly along the top of the frill.
Detail of pin-tucks in the flounce of the original petticoat.
TOP FLOUNCE
If using the width of the fabric, join sections together using narrow French seams. Leave the centre back seam open at this stage. To make up the top layer of the petticoat, begin by making the pin-tucks in a length of fine muslin. A pin-tuck foot and a twin needle can be used to make this process easier; you will also need two reels of thread. The pin-tucks on the original petticoat are 2mm wide and are sewn in vertical bands around the top layer. Each band is comprised of seven pin-tucks which are all 12.5cm long and when sewn they measure 4cm across. A 4cm gap is left before sewing the next band of pin-tucks and this is repeated all around the top layer. The band is 16cm long and so the space below the pin-tucks flares out allows the layer below to be cut from a longer piece of fabric.
The first layer of the reproduction petticoat is also 16cm long and has just two pin-tucks which are 12cm long, sewn every 6cm. On the original petticoat the layer below the pin-tucks is a row of lace followed by a layer of muslin; this is repeated until three layers of lace are inserted. The final layer of the flounce is a frill edged with a delicate picot lace if this can be sourced; an alternative is to use a machine embroidery finish. The top layer of pin-tucks is joined to the petticoat with a row of eyelet lace, which would have been threaded with satin ribbon.
Inserting lace in the flounce.
The reproduction petticoat has two rows of insertion lace. The first is sewn to the bottom of the pin-tucked section by placing the lace on the right side, along the bottom of the pin-tucked section with a 1cm overlap. Pin the lace in place and machine along the edge of the lace. The other layers of fabric and lace are sewn by the same method. To neaten the fabric behind the lace roll over twice and topstitch through all layers or trim the raw edge down to 0.5cm and let the raw seam allowance sit behind the lace.
Preparing the frill to sew around the edge of the flounce.
To finish the flounce a layer of cotton lace is sewn on top of the pin-tucked layer. Place the top edge of cotton lace along the flounce position thread marked line traced, and pin and tack. If possible, check the length is even all around by placing on a dress stand before machining the lace in place, sewing close to the top edge. The lower frill around the bottom of the flounce is made in the same way as the previous frill, although the bottom edge has been neatened by turning twice and using a decorative stitch on the sewing machine that forms small scallops.
ADAPTING THE PETTICOAT
To make a simpler petticoat suitable for a woman on a lesser income the top flounce of lace insertions and pin-tucks can be omitted. A final row of running stitch can be sewn 1cm down from the top of the frill; this will secure the frill in place. Many cheaper petticoats have a frill sewn using the same method as the frill sewn around the bottom of this petticoat. Alternatively the flounce can be reduced in width and made without the lace insertions. If a fuller petticoat with extra volume is required, the width of the flounce, before gathering onto the crocheted band, can be increased. An additional layer of flounce can also be added underneath to increase the width of the hem of a skirt.
Chapter 6
Blouse with Tucks and Lace Insertions | 6
---|---
The Morning Leader _printed an article in 1907 concerning two sisters described as 'decayed gentlewomen'. Having fallen on hard times the two women applied for blouse-making outwork at a Regent Street warehouse. Fabric was supplied, but barely enough to do the job and so cutting out the blouse was 'a horrible sort of nightmare'. The article explains, the blouse had 'hundreds of little pin-tucks' and was the sisters' own design. They worked together all one evening and then all day and late into the next evening to sew the blouse. They had to use their own buttons, sewing silk and ribbons, which was something most dressmakers would have been expected to do. The shop was said to have been delighted with the finished result. From the initial conversation at the warehouse the women had expected to be paid eight shillings and six pence but in fact were paid just six shillings between them. This was a common practice for home dressmakers, who were not covered by employment laws. The disturbing conclusion to the story is that two days later the blouse was displayed in the shop window of the establishment, fraudulently advertised as 'The Latest From Paris' with a price of six and a half guineas, a handsome profit for the unscrupulous warehouse._
The reproduction blouse mounted on a black dress stand.
The blouse featured in this chapter is from Worthing Museum, museum accession number 54/529. At first glance this might look like a difficult project to reproduce but it is actually a basic shape and all pin-tucks and lace sections are made separately and then sewn on top of each other using the pattern as a template. The original blouse is entirely sewn by hand with small, neat and regular stitching, which was obviously carried out by a skilled maker. The blouse has a small square woven label at the centre back waist. The red number 36 indicates the bust size and careful measuring of the garment verifies this to be the case. A size label suggests that the blouse was made to be sold in a department store or small shop, rather than it being a garment made for an individual customer. This makes the fact that it was hand sewn all the more remarkable. Hopefully the maker of this blouse fared better with her employer than the sisters described in _The Morning Leader_ article.
Lace blouse with three-quarter-length sleeves, high neck and pouched bodice sketched by fashion illustrator Ida Pritchard, c.1906. (Worthing Museum and Art Gallery)
Front and back view illustrations of the original blouse from Worthing Museum and Art Gallery.
The blouse is made from fine white cotton with decorative tucks, pin-tucks and lace insertions. It has a centre front panel of hand-embroidered cut work that is edged with a wide lace insertion. It has a boned collar made from straight lengths of lace, and is bound at the top edge with a narrow strip of cotton fabric. The three-quarter-length sleeves have two rows of insertion lace running down the centre of the sleeves, which sit either side of a hand-embroidered panel with cut work. There are three sets of tucks at either side of the insertions. The sleeves are finished with cuffs made from three straight pieces of lace joined together in the same way as the collar; the cuffs are also finished with a narrow binding of cotton fabric. The front of the blouse is longer than the back to allow the front to create a pouched effect. The blouse fastens at the centre back with metal snap fasteners and two shirt buttons and hand-worked loops on the collar. It would have been extremely difficult for a woman to get dressed in this blouse without help. A lady's maid would have been needed to close the fastenings.
CUTTING
Cutting and making this blouse is a bit like doing a jigsaw puzzle: it builds up one piece at a time. All of the tucked sections are straight pieces of fabric that are shaped to the pattern once the tucks have been sewn in place. The central panel forms the template for a layer of insertion lace that is sewn around it. This means that the basic pattern cannot be used to cut out any sections until the decorative pieces are formed and assembled. It also means that there is an element of artistic licence in assembling the blouse and additional insertions of lace can be added if required.
The first section to prepare for cutting is the centre front panel. Mark the straight grain of the centre of the embroidered panel with a row of tacking, and then centre the template over the panel. Pin in place and tack carefully around the edge. Add 1.5cm seam allowance all around and cut out. The sleeves can be left plain and cut from the fabric or they can have rows of insertion lace added down the centre. The completed section is placed on top of the pattern and the outline is traced with thread marking. A seam allowance of 1.5cm is added before cutting out. The lengths of lace for the collar and cuffs can be cut by using the pattern pieces as a guide.
MATERIALS AND EQUIPMENT
2 metres white cotton voile
5 metres × 3.5cm crocheted lace insertion or straight edged cotton lace with an open weave
2.5 metres × 2cm straight edged cotton lace with an open weave
Centre front cut work embroidered panel
2 mother-of-pearl flat shirt buttons
50cm plastic boning (such as Rigelene) for the collar
15 metal snap fasteners
Austrian cotton faggoting (optional)
Sewing machine with pin-tuck foot
Double sewing machine needle
2 reels of white thread
Dress stand
Pattern Master
Air erasable marker
Blouse pattern.
Making order
FRONT PANEL
Insertion lace was extremely popular in the Edwardian era; it does not enjoy the same popularity now and therefore is difficult to source. I discovered some handmade scalloped edge lace in a similar pattern in an antique shop. I washed the lace and placed the scallops so that one edge sat on top of the other. Once the scallops had been zigzagged together and pressed flat they made a good wide insertion lace. A further difficulty was the hand-embroidered panel. To recreate this effect I have used an embroidered table runner from the same period. Many small cloths and runners are still in existence in antique and vintage shops. Once the panel has been thread marked and cut using the template, cut the lace insertions that outline the centre panel leaving enough seam allowance on the end of each piece for joining. Pin the lace insertion on top of the front panel all around the edge, with the edge of the lace sitting alongside the tacking. Topstitch close to the edge nearest to the front panel.
The front panel marked out with pins and the first layer of lace sewn down the sides.
Turn the panel the wrong way round and using the Pattern Master and an air erasable marker, mark 1cm away from the stitching. Cut along this line and then press the edge over by 0.5cm. Press again until all raw edges are neatened, pin and tack in place and slipstitch along the folded edge. The insertions on the original blouse are joined by hand using an overstitching or overcasting method.
The front neck is filled in with three straight rows of insertion lace that are shaped into a V at the front neck by forming a dart on the reverse. Each layer is attached separately and the dart made before moving onto the next layer. To add the first layer take a straight piece of lace and pin it slightly under the top lace attached to the front panel. Make sure the centre will pattern match when made into a dart. Machine the lace in place by topstitching and press. To form the dart lay the section face down on a flat surface and pinch the excess lace into a dart. I found it easier to sew the dart by hand because it makes it easier to control the matching of the pattern. Repeat this process until all three layers are in lace, making sure that the pattern in each layer mirrors the pattern in the layer below.
Back view of the front panel showing how to neaten the front panel by folding the seam allowance twice and slipstitching along the fold.
Back view detail of the front panel showing where to form a dart in each layer of lace sewn above the front panel before adding the next layer.
PIN-TUCKS AND TUCKS
Add a pin-tuck foot and double needle to the sewing machine and follow any guidelines in your machine instructions for stitch length and tension. I used a small stitch and did not adjust the tension. Thread the machine with two reels of thread, one passing through each needle. It is advisable to make a sample before attempting to sew pin-tucks on the blouse.
To make the side bodice sections cut a square of fabric 80cm × 80cm. Use the completed front panel as a guide to mark a curved line to show the position of the pin-tucks, although it is a good idea to continue sewing the tucks 2cm up beyond this line to allow for any movement in the fabric. Measure 6cm in from the selvedge and draw the first line with an air erasable marker. Using the guide of the pin-tuck foot, machine the first row of pin-tucks. Each line after this should use the groove of the foot as a guide to follow the previous pin-tuck. Press lightly once all pin-tucks are sewn and pull the fabric back into shape.
Add a double needle and pin-tuck foot to the machine and begin sewing the pintucks.
Use a single needle and the usual machine foot to sew rows of tucks alongside the pin-tucks.
Thread mark around the edge of the pattern to show the seam lines and add seam allowance.
Completed pin-tuck sections with the position of the front panel marked with the purple lines of an air erasable marker.
To join the front panel to the side panels, place the tucked sections on top of the pattern, lay the front panel on top and pin together.
To sew the tucks, replace the pin-tuck foot with the usual machine foot and remove the double needle and replace with a single needle. Use an air erasable marker to mark a straight line alongside the last pin-tuck. The line should be 1.5cm away from the pin-tuck. Mark 14 further lines 2.5cm apart. To sew the first tuck, fold along the line and press; the tuck is machined 0.5cm in from the folded edge. Continue machining one tuck at a time until all tucks are sewn in place and then press away from the centre front. Use the pattern as a template to place the lace panel over the tucked section. To cut out the shape of the blouse place it over the pattern and thread mark around the edge. The back panels have vertical rows of tucks with a section of pin-tucks in the centre. Above the pin-tucks there is an insertion of lace.
SEAMS
The side seams are offset and sit towards the back of the blouse. The seam is joined together with a readymade insertion of Austrian faggoting. Place the faggoting with right sides facing on top of the thread marked seam line. Machine close to the right side of the faggoting and then trim the blouse seam down to 0.5cm. Pin and topstitch the blouse to the outer edge of the faggoting trim and press. On the original blouse the seams have been rolled and overcast. The armholes in the original garment feature an insertion of cotton faggoting which has not been added to the reproduction.
Lay the centre of the faggoting on top of the seam line and machine down the right side.
There is no shoulder seam on the original blouse. The front and back sections are joined at the shoulders either side of a row of 3.5cm insertion lace.
CENTRE BACK FASTENING
The original blouse fastens down the centre back with 15 metal snap fasteners which are fixed onto strips of satin ribbon. This was a readymade closure which is no longer produced. A contemporary substitute is readymade plastic snap fasteners attached to cotton tape; alternatively, individual press studs can be sewn to satin ribbon or cotton tape.
SLEEVES
The sleeves on the original blouse have two rows of insertion lace down the centre of a hand-embroidered panel; this can be achieved by using the pattern as a template. The reproduction blouse has one layer of 3.5cm insertion lace. To add one layer of lace, cut the lace to the length of the sleeve. Tack a straight line from the tip of the sleeve head to the bottom edge of the sleeve. Centre the lace along the line and topstitch close to the edge of the lace. On the wrong side cut up the centre of the fabric under the lace and roll the seam allowances over and pin in place and either slipstitch by hand or topstitch through all layers. To add tucks either side of the insertion lace in the sleeves, cut straight pieces of fabric measuring 45cm long and 22cm wide. Mark three tucks down the length of the fabric, each 2.5cm apart. The tucks are then machined in place by sewing 0.5cm in from the fold. The insertion lace should then be added and the tucks pressed away from the lace. Place the newly assembled panel over the sleeve pattern and thread mark around the edge of the sleeve and add a 1.5cm seam allowance. Before inserting the sleeve, using a long stitch, machine and sew two rows of gathers between the balance marks along the sleeve head and along the bottom of the sleeve, within the seam allowance. To insert the sleeves, with right sides facing, match up the balance marks and pin and machine in place. The seams can be trimmed and either overlocked together or rolled and overcast.
CUFFS
The cuffs are formed from three layers of straight lace joined together, laying the two outer layers over the wider inner layer. The lace is joined by machining the layers together with a straight stitch. There is no opening in the cuffs; they should be loose enough to pass over the hand. A cotton binding finishes off the edge of the original cuff although this has been left off the reproduction blouse. To add a binding cut a strip on the straight grain and place against the outside of the cuff. With the edges lined up, machine a narrow seam. Fold the binding to the reverse side of the cuff and fold twice to neaten before slipstitching in place. When the cuffs have been added the sleeves can be joined using a French seam.
Overlap the two outer layers of lace over the inner layer to form the collar and cuffs.
COLLAR
The collar is formed from three straight rows of cotton lace using the same method as the cuffs. The original lace may have been crocheted. The open weave of the lace allows the collar to stretch and mould to the shape of the neck. In total the original collar is 7cm deep which includes the flat binding along the top. The collar fastens with two flat mother-of-pearl buttons and two hand-worked loops at the centre back. There are two sections of collar boning positioned at angles to the centre front. The boning feels like steel rods and is covered in white satin fabric and has flossing at the top and bottom. The satin is machined down the sides and given that this is a handmade blouse, indicates that the boning was purchased readymade. A reproduction set of boning can be made by covering plastic boning with readymade satin bias binding. Because of the way the blouse has been constructed the neck edge is not rounded and so it is easier to judge the position of the collar by placing the blouse on a dress stand. Mark the centre front of the collar and line it up with the centre front of the neck. The collar sits on top of the blouse and once pinned in place is sewn on by hand.
Place the blouse on a stand and pin the collar around the neck.
ADAPTING THE BLOUSE
The basic blouse shape can be used for a simpler version although even the simplest of blouses had some form of decoration, for example blouses at the cheaper end of the market had the insertion lace sewn on top of the fabric. For a blouse suitable for a woman on a lower income in the Edwardian period the basic blouse shape could be made using a small floral print fabric. Snap fasteners are useful for a quick change and could be used instead of buttons on the collar; alternatively, circles of Velcro could be sewn down the centre back.
Chapter 7
Two-Part Walking Dress | 7
---|---
_The plain skirt and bodice, also known as a tailor-made costume, was a useful and versatile element of the Edwardian woman's wardrobe. The skirt could also be worn separately with a blouse. In 1900 'Madame Modish' wrote in The Lady's World of the practicalities of a plain skirt: 'we... find it very convenient to have our skirts adorned without introducing a foreign element that might quarrel with one blouse while it fraternised with another.' The walking dress chosen for this chapter comes from Worthing Museum and dates from c.1901, museum accession number 1962/1627/1 &2. It is made from a loose-weave grey wool, which is mounted onto Silesia. Curator Gerry Connolly suggests that this walking dress is the type of outfit worn by a governess, and it has many features which support this idea. It is a practical and plain outfit but has many delightful details. The bodice is fastened down the centre front with fifteen small black flower-shaped shank buttons. There are two further buttons, one on each cuff. It has a straight stand collar that is trimmed on the inside with a band of gathered ivory lace that has been loosely hand stitched, indicating that it was removed for laundering. The sleeves have a false cuff sitting under a sleeve which flares out at the lower edge. The bodice is fitted to the body and has pleats either side of the centre front and three sets of tucks machined in place either side of the centre back._
The reproduction two-part walking dress.
Front and back detail of the walking dress.
Around the waistline of the bodice are four evenly spaced flat buttons used for fastening the bodice to the skirt. The waistband of the skirt has four horizontal buttonholes and a cross, stitched in red thread at the centre front, making it likely that a belt would have been worn on top to hide these details. The skirt is comprised of five gored panels with a centre back, placket opening. It has three useful pockets including a decorative front patch pocket, which is well constructed and hand sewn to the skirt with two rows of stitching and so would have been strong and functional. A tiny square pocket is sewn at the front of the inside of the waistband, which might have been used for a watch. The final pocket is concealed in a back pleat and the purpose of such a pocket was explained by Cynthia Asquith in her memoirs, _Remember and Be Glad._ Reflecting on herself as a seventeen-year-old debutante in 1904, she did not recall when she first carried a handbag, rather she used a 'placket-hole' pocket in her skirts for carrying a leather partitioned purse and a handkerchief. This is the only example of a plain wool skirt that I could find in either museum. One explanation might be that skirts were made from large pieces of fabric and therefore could have been unpicked and remodelled at a later stage. There are many examples of decorative skirts at both museums. Royal Pavilion & Museums have a gored skirt in a wide black-and-white stripe with a flounce at the hem topped with a broad zigzag of black braid. Worthing Museum even has a collection of simple lightweight striped summer skirts still bearing the shop label and prices. A medium-weight linen has been used for the reproduction dress, which has a similar weave to the original wool and is a close colour match, although if a similar wool fabric can be sourced this would be a better fabric for a costume due to the creasing properties of linen. The skirt is mounted with black cotton and the bodice with a striped grey and cream vintage cotton lining fabric. Cutting and making instructions have been given separately for the skirt and bodice in case readers prefer to make only the skirt. In each case enlarge the pattern, add seam allowance and use it to cut and make a toile to test the shape and fit of the dress before cutting in fabric.
MATERIALS AND EQUIPMENT
5 metres grey loose weave wool
3 metres Silesia or similar cotton for mounting the skirt
2 metres striped cotton lining for mounting the bodice
2 skirt hooks and bars
4 snap fasteners
Collar canvas
Thread
17 small shank buttons
Petersham for the waistband
Sewing machine
SKIRT
Cutting and mounting
Begin by cutting the mounting sections. Press the fabric, fold in half and lay the skirt pattern pieces on the mounting fabric by following the grain lines on the pattern pieces. Pin in place and use tailor's chalk with a sharpened edge to draw around the edge of the pattern; add seam allowance and cut out. Remove the pattern pieces and use carbon paper and a tracing wheel to transfer the tailor's chalk pattern markings to the under piece of mounting fabric.
Walking dress pattern A.
Press the skirt fabric, and, if using wool, shrink the fabric by hovering the steam iron close to the surface and then allowing to dry. Fold the fabric in half lengthways with right sides together and the selvedges matching, and place all pattern pieces on the fabric with the grain lines following the straight grain of the fabric. Pin in place and add a generous seam allowance of at least 3cm all around and cut. Remove the pattern pieces from the fabric. Separate the skirt pieces and lay each one flat on the cutting table with the wrong side facing upwards.
The original placket hole pocket is not reinforced with interfacing but this is something to consider at this stage. A rectangle of fusible interfacing can be placed over the pocket opening on the reverse side of the fabric before mounting takes place.
Walking dress pattern B.
Place the corresponding mounting fabric piece on top of the skirt fabric piece with the pattern markings facing upwards. Pin in place, taking care to smooth both layers until flat. Using a single thread, tack along the pattern lines through both layers. Make sure the top corners of each skirt piece are clearly marked on the right sides. Also thread mark the placket opening, balance points and pleats in the two back pieces. Use the hem facing marked along the bottom of the skirt pattern to cut the hem facing pieces from the skirt fabric; add seam allowance.
Making order
PLACKET HOLE POCKET
According to _Weldon's Home Dressmaker_ the placket hole pocket must always be positioned on the left-hand side of the back of the skirt. This is most likely because it was easier to reach with the right hand. It is easier to make the placket hole pocket before the skirt is sewn together.
The pocket bag is a long rectangular shape with rounded edges. Place the pocket facing over the centre of the pocket bag, turn the edges of the facing under and topstitch along the folded edge of the facing. Mark the centre of the pocket from the top to the bottom of the short edge of the bag with a row of tacking. Before attaching the pocket, machine an elliptical shape around the opening on the skirt using a small machine stitch; this will help to reinforce the pocket opening. Working from the right side of the skirt, place the pocket bag over the opening with the right side facing upwards. Pin in place and working from the inside of the skirt tack a line through all layers to show the position of the elliptical line of reinforced stitching. Turn the skirt with the right side facing you and machine just slightly outside of this line. The opening now needs to be cut open using small, sharp scissors. It is a good idea to start in the centre and cut out towards each edge – this minimizes the risk of cutting through the stitching.
Walking dress pattern C.
The pocket bag is joined together using a French seam, which sits on the outside when finished. To join the pocket bag place right sides together and pin around the edge. Machine along the seam line and trim the seam allowance down to 0.5cm. Clip into the seam at the rounded corners and push the bag through the pocket. Press along the edge. Machine a second seam, which will enclose all raw edges. The raw edges of the pocket can be covered with bias binding which is hand sewn in place for a neater appearance. The original skirt has a small piece of cotton tape sewn between the top of the pocket bag opening and the waistband for reinforcement.
SEAMS
Join the seams of the skirt by placing right sides together and pinning through the seam lines. The edges of each seam can then be overlocked and pressed open. On the original skirt the seams are finished by overcasting and are then pressed together to one side, away from the centre front or centre back.
Detail of the back of the skirt showing the placket hole pocket and centre back fastenings.
Inside skirt detail showing the bias tape sewn on top of the raw edges of the placket hole pocket.
Bottom of the centre back placket.
CENTRE BACK PLACKET OPENING
The skirt opens at the centre back and the opening is finished with a concealed placket. The original skirt has evenly spaced hooks and eyes sewn down the placket; the reproduction skirt has snap fasteners. To make the placket, first mount the placket with Silesia and make sure the seam lines are visible on both sides. Press the seam allowance inwards along one long side. Fold the band in half lengthways with the right side facing outwards and press. Snip into the seam allowance at the bottom of the placket opening. On the right side of the skirt place the unpressed side of the placket down one side of the opening, matching the seam lines. Pin in place, stretch the opening to lie flat and pin the placket up along the other side. Machine the placket to the skirt opening, stopping at the midpoint to lift the foot of the machine and pivot the skirt before sewing up the next side. Trim and grade the seam allowance and press along the right side of the newly sewn seam. With the raw edges of the seam enclosed, pin the free edge in place to sit just a fraction to the outside of the previous seam, and pin and tack in place. Working on the right side of the skirt, sink stitch to secure the placket in place. The placket folds into the skirt and the midpoint can be machined in a triangle to finish. Fold the placket so that when you look at the skirt the right-hand side is folded into the skirt and sits on top of the left-hand side.
THE WAISTBAND
The waistband is mounted with Silesia and is stiffened with a band of Petersham. With right sides facing, place one long side of the waistband around the edge of the waist, and pin in place through the seam lines. Tack in place and test the fit of the waistband on the wearer; adjust if necessary and then machine the waistband to the skirt along the seam line. Trim and grade the seam and push the waistband downwards and out of the way. Working from the inside of the skirt, pin the bottom of the Petersham to the seam, just above the line of stitching and machine along the lower edge of the Petersham. Bring the waistband up behind the Petersham and fold it over so that the right sides are facing. Pin down the short ends of the waistband and machine in place. Snip off the corners and turn the right way round and fold the waistband into place. With the long raw edge turned under, pin the waistband to the skirt. Either slipstitch the waistband to the skirt, or working from the right side of the skirt, sink stitch the waistband in place (this process is also called 'stitch in the ditch'). Once the waistband has been completed, cotton tape hanging loops measuring 7.5cm on the double are attached to the original skirt using a cross stitch made with strong thread. Longer loops may be required for storage in a theatre wardrobe. Two skirt hooks and bars are sewn to the edge of the waistband to finish, using a double thread and buttonhole stitch.
Inside view of the waistband showing the Petersham machined to the skirt seam allowance.
THE WAISTBAND POCKET
This tiny pocket is made from a single rectangle of the skirt fabric measuring 12cm × 6cm plus seam allowance. It is folded in half widthways, seamed and turned through. The pocket is then sewn by hand to one side of the centre front inside the waistband. A small pocket watch or locket can be placed inside for safekeeping.
Detail of the inside of the original skirt showing the small watch pocket.
PATCH POCKET
To make the patch pocket, with right sides facing join the pocket lining to the pocket along the straight edge leaving a gap of 6cm for turning through. Press the seam and press lightly along the fold at the top edge of the pocket. With right sides facing, pin around the edge and machine along the seam line. Trim the seam allowance and snip off the top corners and clip around the curved edge at 1cm intervals. Turn the pocket the right way round by pulling through the opening. Use a bamboo point turner to push the corners out and to make sure the edge is crisp. Press lightly and then close the gap by slipstitching by hand. The pocket is loosely gathered 3.5cm down from the top with a row of a running stitch made by hand. Thread a needle with a double thread matching the colour of the fabric and sew a row of even running stitches. Pull to lightly gather and back stitch at the other end. Make four small tucks at the bottom of the pocket, 1cm apart, from the centre front. Secure at the back by backstitching. The patch pocket is attached to the skirt by hand. Place the skirt on a dress stand if possible and position the pocket 6.5cm below the bottom of the waistband and centre it over the right side front seam. Pin in place, and the pocket can then be slipstitched around the edge. For a more secure pocket, pin the pocket in position leaving the area around the edge of the pocket free. Lift up the edge and use double thread pulled through beeswax to slipstitch the pocket to the skirt using small stitches 1cm in from the edge. To finish, use a single thread to slipstitch the pocket around the edge.
Add lining to the patch pocket.
Position the patch pocket on the outside of the skirt.
THE HEM
Skirts observed in museum collections for this book are all constructed to have hem facings. The facing used on this skirt is a separate, shaped piece of fabric 10cm deep, which is sewn to the inside of the skirt hem. The reason for a separate facing could have been that it would be easier to replace a facing rather than a whole lining if the hem became damaged from trailing along the ground. Brush braid or plain wool braid was an additional feature observed in some skirts. Brush braid was a mohair brush edge sewn around the inside hem of the skirt and it must have been especially useful for skirts with a train. The skirt pattern is marked 10cm up from the hem to give the patterns for the facings. Each facing is cut separately from mounting fabric and then joined at the side seams. The seams should be pressed open and the seam allowance along the top edge of the facings pressed under and snipped at intervals to the seam line. To join the facing to the skirt, place the facing over the bottom edge of the skirt with right sides together. Line up all seams, pin along the bottom edge and machine along the seam line. Trim the seam by grading the seam allowance. This eliminates bulk and helps it to sit better. Understitch the lining to the seam allowance and press the hem upwards, on the inside of the skirt. Pin in place and slipstitch, or herringbone if preferred. The original skirt is overcast using large stitches.
Detail of the original hem showing the facing with wool braid sewn along the bottom edge. (Worthing Museum and Art Gallery)
BODICE
Cutting and mounting
Press the fabric and if using wool shrink by hovering the steam iron close to the surface to shrink; then allow to dry. Fold the fabric in half lengthways with the selvedges matching and place all pattern pieces on the fabric with the grain lines following the straight grain of the fabric. Pin in place and add seam allowance. Pattern markings need to be transferred onto the wrong side of the centre back panels and the wrong side of the centre front panels so that the pleats and tucks can be sewn in place before mounting.
Before mounting the bodice with the striped lining, the tucks at the centre back need to be formed. There are three tucks at either side: each one is 0.5cm deep and they are machined in place and pressed away from the centre back. The original garment has a line of machining at waist level over the tucks, as indicated on the pattern. This is an additional process to help keep the tucks lying flat and facing in the same direction. The process was not done on the reproduction bodice.
The front panels have three pleats at either side of the centre front opening. The pleats are machined from the lower edge to the waist on the reverse side of the bodice and then working from the right side the pleats pressed towards the centre front and tacked in place to make mounting easier. A dart on each side of the bodice is also closed before mounting. To close the darts, working on the wrong side of the fabric, fold the dart down the centre, pin in place and machine from the wide end to the point and press towards the centre front. At this stage close the corresponding dart in the front bodice mounting fabric.
To mount the bodice sections place the right side of each outer section face down on a flat surface and place the corresponding mounting pieces on top. When all pieces are flat and smooth, pin in place following the seam lines marked on the mounting pieces using a single thread to sew a row of tacking along each line. Mark balance points such as the shoulder point at the tip of the sleeve head and the sleeve front and back balance points.
Making order
Begin with the centre back seam. Place right sides together, pin through the seam lines, machine and press the seam open. The back side panels have curved seams; with right sides facing, pin together along the seam line starting from the lower edge with the smaller of the two pieces facing towards you. Press your thumbs into the seam and curve it in an arc to help the seams fit together. Place the pins across the seam as well as in the seam and place under the machine foot with the shorter section on top. Machine slowly and lift the foot and pivot the needle when necessary to get a smooth curve to the seams. Press the side back seams together, away from the centre back. The next stage is to join the shoulder and side seams together by placing right sides together, pinning and machining along the seam lines and pressing the shoulder seam open and the side seams towards the centre back. The seams on the original jacket have been trimmed down to 1cm and overcast separately; on the reproduction the seams are trimmed to 1.5cm and overlocked together. If a heavier wool fabric is used for the bodice then it would be advisable to overlock the seams separately.
Mount the centre back panels once the tucks have been machined in place.
Detail of the inside of the original bodice showing the button band and the position of the boning channels with yellow flossing stitches at the tips.
The original bodice is boned around the waist. Two boning channels 16cm high are sewn to the front of the mounted section before mounting takes place, and the boning is inserted into the channels once the bodice has been made up and before the hem facing is sewn in place. A further boning channel of the same length sits over the seam at the centre back and is machined to the seam allowances. The other seams may have been boned when the bodice was made.
Finish the bottom edge of the bodice with a strip of fabric 2cm deep plus seam allowance and 75cm long. Pin the band to the outside of the bodice and machine along the seam line. Trim and grade the seams and press the band over to sit on top of the lining. Fold the top raw edge under and slipstitch to the lining.
The bodice is closed at the centre front with shank buttons and buttonholes. The button bands can either be mounted with Silesia to replicate the original bodice or a layer of fusible interfacing can be applied to the reverse side for extra strength. Fold each button band in half lengthways with right sides facing and seam across the short edges. Snip the corners, turn the right way round and press. With right sides facing and raw edges aligned, position a band along the centre front edge. Pin and machine along the seam line. Trim and grade the seam and push all raw edges into the band. Fold the remaining raw edge under and either slipstitch in place from behind or sink stitch by machining on the right side of the bodice. Position the buttonholes down the right-hand section and ensure that when the buttons are closed they will sit in the middle of the band. The first buttonhole sits 1cm down from the upper edge and the other buttonholes are spaced 2.5cm apart. Follow instructions on your machine for sewing buttonholes. Alternatively buttonholes can be sewn by hand using buttonhole twist to match the original bodice.
MAKING THE SLEEVES
Begin by making the false cuffs, which are not lined or mounted. The pleats radiate outwards from the centre of the top and bottom seam lines. Pin the pleats in place following the directional arrows shown on the pattern and machine within the seam allowance. With right sides facing, join the seams down to the top of the opening by placing right sides together, pinning through the seam lines and machining and pressing the seam open. Overlock the edges of the seams and fold the cuff opening back twice and machine to a point above the opening. The lower edge is finished with a band that runs around the wrist. To make the cuff bands first mount each piece with Silesia. Fold each cuff band in half lengthways with right sides facing and seam across the short edges. Snip the corners, turn the right way round and press. With right sides facing and raw edges aligned position a band along the edge of the pleated cuff. Pin and machine along the seam line. Trim and grade the seam and push all raw edges into the band. Fold the remaining raw edge under and either slipstitch in place from behind or sink stitch by machining on the right side of the cuff. Machine one buttonhole running horizontally and a corresponding button.
Make the false cuffs into cylinders.
The next stage is to make up the lower sleeve lining. The facing running around the edge of the lining is a separate piece, which has been marked on the bottom of the pattern piece. Alternatively, wide bias binding can be used. Bias binding can be shaped to fit a curved area by stretching and manipulating it into shape with the iron. Thread mark the seam lines on the facing, snip into the curves at 1.5cm intervals and press the top seam allowance under. Place the facing on top of the sleeve lining, pin in place and topstitch along the fold.
Add the facing along the bottom edge of the lower sleeve lining.
With right sides facing, join the seams of the lower sleeve lining and press open. There is no need to finish the edges of the seams because they will be concealed in the jacket when it is made up. With right sides facing, slip the lining over the false cuff making sure the seams are sitting on top of each other. Tack both sections together along the upper seam line. Join the upper lining seams together by placing right sides together, machining in the seam lines and pressing open. With right sides facing slip the upper sleeve over the lower cuff, lining up the seams and with raw edges also lined up. Pin through all seam lines and tack in place. Machine around the seam lines and trim and grade the seams. Press the seams upwards towards the sleeve head.
The next stage is to join the outer sleeve to this section around the lower edge. With right sides facing, pin all around the lower edge and then machine within the seam line. Trim the seam and snip at intervals. Turn the right way and press along the edge for a crisp finish.
Next, the lining and sleeve need to be mounted together at the sleeve head and the seam lines and balance points tacked through. Two box pleats are folded into place at the sleeve head by following the directional arrows on the pattern. Machine the pleats in place within the seam allowance.
SETTING IN THE SLEEVES
To join the sleeves to the bodice, work from the inside of the sleeve and match the balance points around the sleeve and the bodice, smoothing the sleeve into the bodice as you go. Working from the inside of the sleeve, pin vertically and horizontally and tack. Check the position of the sleeves by placing the bodice on a dress stand before machining within the seam line. Trim and overlock seams together. On the original bodice the seams are overcast separately and pressed into the sleeve.
Insert the sleeves.
THE COLLAR
Cut a piece of collar canvas larger than the pattern piece and trace around the edge with tailor's chalk. Place on the reverse side of the under collar, and with both sections lying on a flat surface, pin in place and tack through all layers. Machine the collar canvas along the seam line. Using a small, sharp pair of scissors, trim the seam allowance of collar canvas away, cutting close to the row of stitching. With right sides facing, pin the top collar and under collar sections together. Machine the sections together along the seam line; when passing around the corners use a small stitch and lift the machine foot and pivot to turn the collar. (Some costume makers prefer to sew one diagonal stitch across the point of the collar instead of going into the point.) Trim and grade the seams, snip the corners off, and place a drop of Fray Check at each corner for extra strength. Turn the right way round and push the corners out with the bamboo point turner or similar. Press the collar from the inside and understitch the under collar to the seam along the top as far as possible. Tack around the collar to keep in place. To attach the collar to the bodice, with right sides facing and raw edges aligned, pin the neck edge of the outer collar to the neck edge of the bodice, matching the centre back marks. Pin and machine along the seam line. Trim and grade the seam and snip at intervals. Press the seam into the collar and fold the raw edge of the inner collar under and pin and slipstitch to the bodice. Sew two hooks and bars inside the collar using a double thread and a buttonhole stitch. The collar sits edge to edge when finished.
Apply a drop of Fray Check to the corners of the collar.
Prepare the collar to be attached to the bodice.
Finishing
To finish the collar, gather two pieces of lace together, 35cm × 3cm when gathered, and sew a piece of binding along the bottom. Slipstitch around the top of the inside of the collar so that 1cm of lace is visible.
Detail of the original bodice showing the position of the hooks at the centre front and the lace sewn to the inside of the collar. (Worthing Museum and Art Gallery)
ADAPTING THE WALKING DRESS
The skirt can easily be adapted by the addition of trimmings around the hem, and can be used with a variety of bodices or blouses. This style of skirt was popular until around 1908 when straighter skirts began to appear. The bodice could also be trimmed to make a more elaborate garment. For a quick change in the theatre, the buttons could be sewn on the opposite facing and Velcro sewn underneath.
Chapter 8
Day Dress | 8
---|---
_'To create a beautiful gown is indeed as much a work of art as to paint a beautiful picture' – so claimed the Woman's Institute of Domestic Arts & Sciences Essential Stitches and Seams booklet. The day dress featured in this chapter is from the Katherine Farebrother collection, a sub-collection at Royal Pavilion & Museums, Brighton & Hove, museum accession number C003144. It is a lightweight dress, which would have been ideal for summer days. The dress is comprised of a separate skirt and bodice made from black and cream striped cotton fabric with a jacquard woven raised pattern. The softly pleated bodice is lined but is not boned and pouches over the waistband at the sides. It has a front bib with horizontal tucks made from cream cotton lawn, lined with muslin. Like many Edwardian gowns, the back mirrors the front with details being slightly narrower at the back. Strappings of bias fabric with a row of lace braid decorate the bodice and sleeves. The sleeves are made in two parts: the lower sleeves have rows of narrow tucks and a small opening at the wrist; the upper sleeve is gathered giving the effect of a leg of mutton style. There is a simple inner bodice, which acts as a camisole; it fastens at the centre front with hooks and thread bars. The collar is made from two rows of delicate lace with a floral pattern. The collar is 7cm high in total and is supported by five composite card collar supports, each 5cm high and tacked to the inside of the collar. The bodice has four horizontal hand-worked loops evenly spaced around the waist that correspond with four hooks sewn to the inside of the waistband to help keep the bodice and skirt in place. The skirt is made from six gored panels with a centre back opening. The waistband dips at the centre front and has three bias cut pleats mounted onto a shaped waistband. There is a small pouched bag hanging from the waistband on the right side. Strappings of bias fabric decorate the hem and are finished off with covered buttons. The dress has many fine features but it is not of the highest quality of making; there is no label and so it is possibly the work of one of the local dressmakers Katherine Farebrother was known to employ. A black tie with dangling bobbles finishes the outfit. Instructions are provided in this chapter for making the tie, but it is most likely that it was purchased separately from a vendor of novelties in ladies' neckwear._
The reproduction day dress.
Enlarge the pattern, add seam allowance and use it to cut and make a toile to test the shape and fit of the day dress before cutting in fabric.
Front and back illustration of the original day dress from Royal Pavilion & Museums, Brighton & Hove.
Cutting
For the skirt: press the fabric and lay a single layer of fabric on a large flat surface. Begin with the skirt and the centre back panels. Each panel needs to be cut separately in order to match the centre back stripes and to ensure that they form a chevron when machined together. The centre front panel can be cut so that the centre front line sits on top of a vertical stripe. The side back and side front panels will not match due to the direction of the grain lines on the pattern pieces. The straight side of each panel should be placed on a wider stripe. Once pattern pieces are pinned in place add seam allowance before cutting. Tailor's tacks can be used to mark the centre back opening and the other balance marks. The horizontal pleats on the waistband are made from three bias strips with each strip measuring 7cm × 75cm. Cut three strips of white muslin to the same size to be used for mounting. Cut one piece of fusible interfacing using the waistband pattern. The decorative bands around the hem are cut on the bias and measure 5cm when cut and 2.5cm when finished; they also require muslin backing, which is also cut on the bias. To match the original skirt all stripes must be cut to follow the same direction. A further two rectangles are needed to hang the bag from the waist; these measure 1.5cm to the fold and are 28cm long when completed.
MATERIALS AND EQUIPMENT
5 metres cotton stripe (cotton shirting has been used for the reproduction dress)
2 metres cotton muslin
1 metre cotton lawn or similar
1 metre × 3cm lace for the collar
1 metre edging lace to be gathered on to the top edge of the collar
3 metres × 1.5cm wide straight lace braid
20cm black satin for the tie
2 × 2cm pompoms
Thread
Narrow plastic boning for the collar
50cm satin bias binding
Small hooks and bars
25cm medium weight fusible interfacing
Sewing machine
For the bodice: cut two straight sections of cream cotton fabric measuring 110cm × 30cm for the bib panels at the centre front and back. The bodice has two further front and two back panels and a peplum at the lower back. The tucks on the front of the bodice can be adjusted to follow the stripes of the fabric. The pleats and tucks on the front bodice should be marked with either tailor's tacks or an air erasable marker. The peplum is a single piece of fabric cut on the bias. The sleeves are comprised of an upper and a lower sleeve. Cut a piece of fabric measuring 32cm × 32cm to make the tucked lower sleeve sections. Add a 3cm facing allowance to the bottom of the sleeve; this includes 2cm for the facing and 1cm for turning under to neaten the edge. The front and back bib linings are cut from muslin. The decorative bands on the bodice are cut on the bias and measure 5cm when cut and 2.5cm when completed; they also require muslin backing, which is also cut on the bias. (The original bodice has a separate in-built camisole made from lightweight muslin, fastened with hooks and bars at the centre front; this has not been included in the reproduction.)
Day dress Pattern A.
Day dress Pattern B.
Day dress Pattern C.
DECORATIVE BANDS
To make the decorative bands, also known as strappings, first mount each one with bias cut muslin and press the outer seam allowances into the centre. For the bodice, only apply the mesh braid down the centre once the seam allowances have been pressed into the centre. Even though it seems more aesthetically pleasing to have the stripes forming a chevron, the stripes on the museum's skirt and bodice all face in the same direction. Cover the buttons used as decoration on the skirt with the striped fabric following the guidelines on the packet. Sew the buttons in place with a double thread at the top of each point. The stripes on the original buttons all face downwards. Wrapping the thread tightly around the shank before backstitching helps the button to stay in place and keeps the stripe vertical.
Detail of the original skirt showing the decorative strips applied around the hem with covered buttons at the tips.
BODICE
Making order
FRONT AND BACK SIDE BODICE
Begin by forming the tucks that fall from the shoulders in the front side bodice sections by folding along the lines indicated on the pattern. Pin, machine and press one tuck at a time before moving on to the next. When all tucks are completed press them away from the centre front to face the armhole. The front side bodice pieces have pleats at the waist to create the pouched effect and these are formed by pinching the pleats together on the reverse side and machining a vertical line to hold each one in place. The back panels also have pleats at the waist which can be pinned and machined in place within the seam allowance.
Detail of the bodice of the original day dress open at the left hand side showing the position of the hand worked loops.
Tucks flowing down from the shoulder at the side front.
BIB PIECES WITH TUCKS
To make the bib tucks, measure 2cm down from the top of the cream cotton fabric making sure the line is sitting along the straight grain of the weft thread. Using an air erasable marker draw a straight line, and then draw a further line 0.5cm apart. Fold along the final line and press. Machine 0.5cm down from the fold using the marked line as a guide. Push the tuck away from the top and on the wrong side press flat. Press on the right side and measure 0.5cm from the fold and a further 0.5cm. Press along the outer line and repeat the process until the section is filled with tucks. Alternatively mark the whole section in stripes 2.5cm apart, each stripe being the fold line for a tuck. The tucks on the original bibs are 3mm wide with a gap of 3mm between each tuck. Cut out the front and back bib sections using the tucked panels with all tucks facing downwards. Cut out the same shapes from muslin.
Marking the tucks for the bib using a Pattern Master and an air erasable marker.
Lay the bodice flat and pin the muslin lining around the neck.
Fold the muslin lining behind the bibs and pin to the seam allowances.
JOINING THE BIB AND BODICE PANELS
The front and back bib sections are sandwiched in place before the front and back bodice sections are joined together. Begin by making up the back of the bodice. With right sides facing place the bib to the bodice and pin and machine within the seam lines. Press the seams away from the centre front. Join the right-hand side bodice section to the right side of the front bib, and with right sides facing place the bib to the bodice and pin and machine within the seam lines. Press the seam away from the centre front. Overlock the edge of the left front side bodice and press the seam allowance under along the seam line. Join the shoulder seams by placing right sides together and pinning and machining along the seam line. Press the seam open and overlock the edges.
The neck edge of the bodice is neatened by the muslin lining. Join the shoulder seam on the right-hand side of the muslin lining using a narrow French seam or an open seam. Lay the bodice on a flat surface with the right side facing upwards. With right sides facing, place the muslin lining on top of the bib section. Pin together around the neck and shoulders along the seam lines. Leave the left-hand side of the front bib open. Machine along the seam line and trim to 0.5cm and snip into the curved edge around the neck at 1cm intervals. Turn the right way round and press before topstitching 0.5cm in from the neckline. Fold the muslin lining to the inside of the bodice and line up all raw edges. Pin and tack in place and overlock the edges; the lining will finally be held in place by the two rows of machining used to attach the decorative strips, which will be added once the sleeves are in place.
The open edge of the bib is finished with a binding made from cream cotton. Cut a bias strip 4cm wide and 56cm long. Fold the long edges into the centre and press. Working from the back of the bib place one edge of the binding along the raw edge and pin and machine down the pressed line. Fold the binding round to the front of the bib, tuck the top end in and pin through all layers of the binding and machine close to the folded edge. A row of thread bars is worked in blanket stitch down the seam line of the left side of the bib, 3cm apart. A row of corresponding hooks is sewn under the strapping on the left-hand side of the bodice when the bodice is completed.
SLEEVES
To make the tucks, measure 6cm down from the top edge of the fabric piece and draw the first line with an air erasable marker. Fold along the line and press. Machine 0.5cm away from the fold to form the first tuck. Press the tuck downwards. Measure 1.5cm from the folded edge and press along the fold line. Machine 0.5cm down from the fold to form the second tuck and machine in place. Repeat until 10 tucks are sewn. Place a decorative strip along the bottom of the fabric with the edge of the tuck butting up against the edge of the strip; pin and machine in place along the edge of the braid. Press the facing over by 1cm and machine along the edge.
Sew tucks in the lower sleeve.
Gather the top sleeve to fit the lower sleeve and add the decorative bands.
Position the pattern piece on top of the lower sleeve and mark around the edge, add seam allowance and cut out. Fold the facing round to the front, stitch down the sides within the seam allowance and snip off the corners. Turn the facing the right way round so that it sits at the back of the sleeve. Slipstitch the facing to the sleeve. On the original bodice the fullness in the upper sleeve at the sleeve head and at the elbow is reduced with tucks that are unevenly spaced. The reproduction sleeve pattern is gathered at the sleeve head and has tucks at the elbow. To make the gathers, set the machine to the longest stitch and sew two parallel lines behind the seam line and within the seam allowance, approximately 0.3cm and 0.6cm apart. Knot the four threads at one end and pull to gather using the top threads at the other end. Check that the measurement fits between the notches marked around the armhole on the bodice and knot the threads at the other end. Distribute the gathers evenly and, using only the tip of the iron, press along the gathers, within the seam allowance. To join to the upper and lower sleeve place right sides together and pin along the seam line; pin and machine together. Trim the seam allowance down to 1cm and overlock the edges together. Lightly press the seam allowance upwards. Place the sleeve on a flat surface facing upwards and pin the decorative band in place along the top of the lower sleeve and just covering the upper sleeve; machine along the edges of the braid to secure.
To join the sleeve seams place right sides together and pin along the seam line. Use pins to match up the decorative bands and tucks. Machine, and overlock the seam allowances and press together to one side. Sew a hook and bar at the wrist so that the cuff overlaps slightly.
With right sides facing pin the sleeve along the seam line.
SETTING IN THE SLEEVES
To join the sleeves to the bodice, work from the inside of the sleeve and match the balance points around the sleeve and the bodice, smoothing the sleeve into the bodice as you go. Pin vertically and horizontally and machine within the seam line, working from the inside of the sleeve. Trim and overlock the edges together. On the original bodice the upper sleeve has a separate and narrower lining joined from the top of the lower sleeve to the armhole and the armhole is encased with a binding to finish all raw edges. The pattern piece can be used to make the lining by folding some of the fullness away.
EPAULETTES AND DECORATIVE STRIPS
The epaulettes are cut on the cross grain of the fabric and on the original garment the stripes face the same direction on both epaulettes. To make, place the pattern on the right side of the fabric, place carbon paper underneath and use a tracing wheel to trace around the edge. Tack along this line to thread mark the outline of the epaulette. The epaulettes on the original garment do not feel as if they contain any interlining. Because the fabric I am using is slightly lighter than the original, I tacked a layer of muslin behind the epaulette to use as an interlining. Beginning with the points, press the seam allowance inwards, then press in the sides. I added an extra 0.5cm to one of the sides and this was pressed under to neaten the edge. The underside is finished off by slipstitching the seam. Centre the lace braid on top of the epaulettes leaving an extra 1cm to overhang at the point. Machine in place down both edges and finish off by folding the braid over the pointed edge and slipstitching in place from behind.
Mount the epaulette with muslin and fold into shape with a pointed end.
Centre the epaulettes over the shoulder seam.
Machine the decorative band to the bodice.
To add the epaulettes to the bodice, centre them over the shoulder seam with the pointed end of the epaulette sitting over the sleeve head by 1cm. Pin and machine in place by sewing along the edges of the lace braid. Place the bodice on a dress stand if possible and centre the decorative strips to sit along the side bodice seam line. At the left front edge the strip will overhang the seam line which has already been folded behind. Machine the strips to the bodice by sewing along the edges of the lace braid.
PEPLUM AND BODICE HEM
Finish the back of the bodice by adding the peplum. To make the peplum, neaten the two short edges and the lower edge by folding the seam allowance over twice and machining. Fold the pleats into the top of the peplum and machine within the seam allowance. Match the centre of the peplum to the centre back mark at the end of the bib and with right sides facing pin the peplum to the bodice. Machine in place along the seam line and overlock all edges together and press the seam downwards. Finish the front hem of the bodice by folding the seam allowance over twice and machining.
Add the peplum to the lower back bodice.
LACE COLLAR
The finished collar on the reproduction garment measures 39cm although this measurement can be adjusted to suit the preference of the wearer. The widest part of the collar is made from two straight pieces of lace with a repeating pattern. The pattern is mirrored on the top and bottom layer of the collar, so to join the pieces together, begin by laying one on top of the other, matching up the pattern. Pin in place and line up the edges. Using a short, wide zigzag stitch, sew along the edges to join together. Open out and press flat. The collar is trimmed at the upper edge with gathered lace, which may have originally been bought as a readymade piece because it is gathered onto a piece of linen that is not used elsewhere in the dress. To make a gathered section of lace, use the longest machine stitch to sew two rows along the bottom edge. Tie a knot in all threads at one side and then pull the top two threads at the other end to gather. Check the measurement against the straight section and knot at the other end. Press the lace lightly using the tip of the iron, staying within the seam allowance. Pin the gathers behind the top of the collar and machine in place. To neaten the front of the collar make a bias strip 2cm wide and 42cm long. Press the long edges into the centre. Place the strip on the right side of the collar and centre it over the join between the top of the collar and the frill. Pin and machine, sewing close to the outside edges. The raw edges behind the collar are enclosed by hand sewing a strip of narrow bias binding on top but leaving the bottom edge open for inserting the collar supports. This edge is hand sewn to the collar supports once they are in position.
Make the lace collar.
The original blouse has five 5cm high composite card collar supports attached at regular intervals around the collar. The supports are loosely hand sewn to the lace collar in order to make removal easy when the collar needed to be washed. Although it is no longer possible to buy composite collar supports, substitutes can be made by covering plastic boning with readymade satin bias binding.
The collar is fastened at the centre back with three thread bars and metal hooks. Fold the short end of the right-hand side of the collar under by 0.5cm and then a further 0.5cm and slipstitch. Thread bars are sewn on top of this seam. To make a thread bar, thread a needle with a double thread and make a knot in the end. Start with a back stitch and make a loop wide enough for the hook and backstitch. Bring the thread to the surface and work a tight blanket stitch along the loop until you reach the other end. Finish with a final backstitch. The three corresponding hooks on the other side of the collar (which is also folded twice and slipstitched) can also be sewn on with a blanket stitch. The collar is attached to the bodice once the bodice has been completed.
NECK TIE
The original neck tie is 8cm wide across the top of the bow and 13cm long from the top of the bow to the bottom of the lower tie. Cut four strips of black satin measuring 6cm × 18cm. With right sides facing, fold in half lengthways and machine down the long edge. Turn two of the strips the right way round and press. Construct the tie by forming a double bow with a section wrapped around the middle. Sew a V shape at the ends of the other two strips with the seam in the centre. Trim around the point and dab a drop of Fray Check before turning through and pressing. These two pieces are used for the dangling sections of the tie and one is 2.5cm longer than the other when they are layered together. A section of the strip is wrapped firmly around the ties 2cm down from the bottom edge of the bow and hand sewn in place from behind.
Assemble the neck tie.
Use embroidery thread to make a chain loop from the lower points of the tie to the top of the bobbles.
To make the bobbles cut two circles of fabric big enough to cover the pompoms. Sew a small running stitch around the edge of the circle, place a pompom in the centre and pull the threads to gather. Finish off by oversewing. There will be a frayed edge and so trim the most prominent threads and cover the area with Fray Check and allow to dry. To attach the bobbles to the ends of the tie use three strands of embroidery thread and make a chain measuring 2cm long, starting at the bobble and finishing with a backstitch towards the back of the ends of the tie. A small safety pin can be sewn to the back of the tie and used for attaching the tie to the front neck where the collar and bodice meet. The original tie is hand sewn to the front of the collar.
SKIRT
Making order
JOINING THE SEAMS
Begin by joining the centre front panel to the side front panels by placing right sides together and matching the balance marks and seam lines, then pin and machine. The seams can then be overlocked together and pressed away from the centre front or overlocked separately and pressed open. The seam allowance on the original skirt is 1cm. The seam edges are unfinished and are pressed together to one side. The side panels should next be joined to the back panels; the centre back seam is the last to be joined and before this the decorative strips should be applied to the skirts. Tack the strips into place and form into a point 18cm high from the tip to the bottom of the band, mitre the corners. The bottom edge of each finished strip sits 8cm up from the bottom edge of the hem. The hem facing, which is cut on the bias, is 10cm deep when finished. The hem facing can be machined to the skirt because the decorative bands will cover this when they are machined around the skirt.
Detail showing the original skirt facing cut on the bias and pleated into place.
CENTRE BACK PLACKET
The skirt opens at the centre back and the opening is finished with a concealed placket with horizontal stripes. The original skirt has evenly spaced hooks and eyes sewn down the placket; the reproduction skirt has snap fasteners. Mount the placket with muslin or fusible interfacing and make sure the seam lines are visible on the wrong side. Press the seam allowance inwards along one long side. Fold the band in half lengthways with the right side facing outwards and press. On the right side of the skirt place the unpressed side of the placket down one side of the opening matching the seam lines. Pin in place, stretch the opening to lie flat and pin the placket along the other side. Cut into the seam at the bottom of the placket. Machine the placket to the skirt opening, stopping at the midpoint to lift the foot of the machine and pivot the skirt before sewing up the next side. Trim and grade the seam allowance and press along the right side of the newly sewn seam. With the raw edges of the seam enclosed, pin the free edge in place to sit just a fraction to the outside of the previous seam, and tack in place. Working on the right side of the skirt, sink stitch to secure the placket in place. The placket folds into the skirt and the midpoint can be machined in a triangle to finish. Fold the placket so that when you look at the skirt the right-hand side is folded into the skirt and sits on top of the left-hand side. Space three sets of snap fasteners equally down the centre of the placket and sew firmly in place using a double thread and a blanket stitch.
Detail showing the original back placket with horizontal stripes.
The top of the elastic casing pinned 3cm down from the top of the bag.
POUCHED BAG
The bag is unlined and is constructed of two pieces joined together at the sides by a French seam that sits on the outside of the bag. To form the seam, place right sides together and machine 0.5cm away from the seam line, trim to 0.3cm, turn the bag inside out and press along the seam line. Sew a further seam 0.5cm in from the edge. Fold the facing to the inside and press along the fold line. Open up and press the raw edge over by 1.5cm. Fold the facing back inside the bag. Mark the casing with an air erasable marker 3cm down from the top of the bag and a further 1cm below this line. Pin and machine the facing to the bag along these lines to form the casing. Leave an opening at the back of the bag for threading the elastic through. Pull the elastic to gather the casing until the bag measures 14cm across at this point. Overstitch the elastic and machine the casing gap closed. To make the ties that join the bag to the waistband, fold the fabric in half lengthways and machine down the side and across one end. Trim and turn the right way round and press. Fold the finished end over by 2.5cm and centre each tie over the side seam and machine in place along the top of the casing.
The finished pouched bag hanging from the waistband.
WAISTBAND
The waistband is shaped to dip at the centre front and gradually rise to a peak at the centre back. It has three rows of folded bias cut strips applied to the front of the waistband, which follow the shape of the waistband. To prepare the bias strips mount each one onto a strip of muslin using long diagonal tacking stitches. Press two of the strips in half with the right side facing outwards. The other strip should be pressed under by 1cm along one long edge only – this is the middle strip. Take the front waistband piece and apply fusible interfacing to the reverse, making sure all seam lines are visible on both sides of this piece. To assemble the waistband, begin by positioning the bottom folded bias strip. This hangs 1cm below the seam line and when the skirt is assembled this section hides the skirt and waistband seam. Pin and tack in place, taking care to keep the bias strip lying flat. Measure up 1.5cm from the folded edge at the centre front and mark a line. Take the strip that has been pressed under by 1cm and position the folded edge along the marked line at the centre front and lift at the centre back. Working from inside, machine along the fold and then press this section upwards. It should hang down slightly and the diagonal lines should match at the centre front. Pin and tack it in place along the top raw edge. Take the third section and position it 2cm up from the previous folded edge at the centre front and swing it out to run along the top of the waistband at the back. Pin and tack in place and machine.
Layer the three long bias strips to the front of the waistband, lifting them at the centre back.
The centre back waistband.
Take the inside facing waistband piece and place on top of the front section with right sides together. Follow the tacked lines, pin in place and machine along the seam line. Grade the seam allowance to reduce bulk and snip off the corners. Place a dab of Fray Check on the corners and allow to dry before turning the right way round. Understitch along the top edge of the waistband as far as possible. Press the waistband to lie flat. Place the pattern piece on top to check the measurements. Add the centre front line and the line to mark the wrap over of the waistband at the centre back. The waistband is attached to the skirt in a back to front way. The inside of the waistband is machined in place first and the front is slipstitched afterwards. To finish, a covered piece of boning is sewn to the inside of the right centre back and three sets of hooks and bars are added.
FINISHING
The collar opens at the centre back and is hand sewn to the neck edge around the right-hand side of the bodice, starting at the centre back and then continuing around the front bib and stopping 2.5cm from the shoulder seam to allow the front bib to slide under the back when the bodice is worn. To finish the collar, sew three hooks along the bottom of the left side of the lace collar and sew three thread bars around the left back neck edge. Place the completed bodice and skirt on a dress stand and pin the bottom edge of the lace collar on top of the neck. The bottom edge of the collar sits 0.5cm away from the neck edge and can follow the line of topstitching. Slipstitch in place with small stitches.
Place the bodice on a stand and hand sew the collar.
ADAPTING THE DAY DRESS
The skirt can be made up in a plain fabric without the decorative trimming. A straight waistband could be attached like the waistband used in Chapter 7. The skirt could also be made up in wool and mounted onto Silesia or a similar cotton backing. For a quick change Velcro could be used as a method of fastening on the placket of the skirt, although it is advisable to use a skirt hook and bar on the waistband. The bodice pattern could also be used to make a blouse, and the necktie could be replaced with an Edwardian brooch.
Detail showing the original collar with composite card collar supports and a row of hooks along the back edge and at the centre back.
Chapter 9
Evening Gown | 9
---|---
_In 1907 'Penelope', a fashion writer for the upmarket weekly journal_ The Ladies' Field, _wrote an article considering the drawbacks of unadventurous evening gowns:_
There is generally a time in a woman's life when she absolutely sighs for a gown which is out of the common – distinguished, artistic. One gets so terribly tired of the ordinary fashionable garb, no matter how prettily it is carried out, and the sure knowledge that a hundred, more or less, exact reproductions of one's self will be present at every crowded gathering, is, after a time, somewhat depressing.
_The solution to her dilemma would have been either to make her own gown or to commission an exclusive design from a dressmaker. The black satin evening gown featured in this chapter is exquisitely beaded by hand-making it unlikely that the wearer would encounter someone else in the same gown at an evening function. The gown is part of the collection at Worthing Museum, museum accession number 1970/802/2. It is listed with an approximate date of c.1918 although the high-waisted column gown also known as the Directoire waist also corresponds to designs from earlier in the decade. A sketch of a black gown c.1912 drawn by fashion illustrator Ida Pritchard has similar features._
The reproduction evening gown.
The dress wraps over at the front and opens on the left-hand side to reveal a foundation bodice and underskirt. The outer body of the dress is in good condition, the satin is unmarked and the chiffon sleeves and detailing are intact. Like many Edwardian gowns the back of the bodice mirrors the front. The bodice has front and back inserted panels of tulle, which are decorated with a geometric design formed from black cord braid with pearlescent sequined flower motifs and silver star-shaped sequins held in place with small gunmetal beads. Below the waist the back of the dress has a small train cut as a separate panel that falls to the left side only. At the centre front the skirt parts to reveal a satin panel sewn to an underskirt. A black satin flower marks the spot where the skirt opens. The dress has been constructed to hang off a cream silk boned foundation bodice; the silk is fragile and has begun to disintegrate in places. Around the inside of the neck of the foundation bodice sits a casing and the original narrow tape drawstrings can still be seen. The purpose of the drawstrings was to ensure a snug fit to the body of the wearer. The foundation bodice fastens with hooks and eyes at the centre front and it is secured to the body at the waist through a woven twill waist stay.
Black Empirestyle evening gown sketched by fashion illustrator Ida Pritchard. (Worthing Museum and Art Gallery)
Illustration of the front and back view of the original evening gown.
Front beaded panel of the original gown.
The waist stay is printed with the details E. & M. Hull, 59 Chepstow Place, Bayswater, and the single W is used in place of West London. This was the manufacturer of the foundation bodice but it may not have been the manufacturer of the evening gown. There is a small hole in the bodice that has been patched and repaired close to the centre front opening, which suggests that the bodice is older than the rest of the dress and may have been used as the foundation of a previous garment. A belt worn at the waist around the outside of the dress is made from a layer of net, heavily encrusted with black glass or jet beads. The belt is lined with satin and fastens at the side front. In 1909 _The Drapers' Record_ reported the popularity of jet cabochon beads used as decoration on evening gowns. One particular example, a black net tabard, was covered with jet cabochons, which made it extremely heavy. The beads used on the Worthing museum dress appear to be a mixture of jet cabochons and glass beads. Large flat single beads are sewn to the top of each shoulder with two further beads used to form narrow horizontal pleats near the hem at the right-hand side. The long floaty sleeves are formed of panels of chiffon with beaded tassels sewn to each of the four lower points. Two further panels of chiffon are gathered onto the waist to form side panels that lie over the skirt. Enlarge the patterns, add seam allowance and use them to cut and make a toile to test the shape and fit of the evening gown before cutting in fabric.
Inside the evening gown, showing the waist stay printed with the name and address of the original dressmakers.
Detail showing the back view of the original evening gown, which mirrors the front view.
Detail of the shoulder of the original gown with a flat bead centred over the shoulder seam.
Cutting
Note the direction given on the skirt pattern pieces for placing the pattern on the fabric. The satin skirt panels should all be placed on the satin as indicated on the pattern to ensure that the train sits on the left-hand side of the body. Both pieces of the underskirt are cut on the fold. A rectangle of fabric and fusible interfacing measuring 6cm × 60cm are also needed to form the placket on the underskirt; this includes 0.5cm seam allowance. A false front for the underskirt is cut from satin following the shape marked out on the underskirt pattern. An additional hem allowance should be added to the bottom edge of the false underskirt panel of 11cm; this includes a 5cm hem and 6cm for facing the hem. The outer bodice is cut in satin and is unlined. The tulle bodice panels have a single layer of black net on top. The foundation bodice is cut from silk habutai. All chiffon pieces should be cut out on a folded piece of fabric following the grain line marked on the pattern. Add a generous seam allowance to all pattern pieces. All seams and balance marks should be tacked through to the right side of the fabric.
MATERIALS
Outer dress and scarf
4 metres black satin
4 × 5cm black beaded tassels
4 × 2cm pompoms
Stranded gold embroidery thread
3 metres chiffon
Underskirt
1.5 metres black cotton fabric
5 black snap fasteners
Foundation bodice
50cm lightweight cream silk habutai
1 metre × 22mm cotton twill tape
1.5 metres narrow cotton tape for the neck casing
Hooks and eyes/bars
Net panels and beading
50cm black net
1 metre lightweight cream tulle
5mm matt black flat sequins
5mm mother of pearl flat sequins
6mm black bugle beads
7mm black glass faceted beads
28g × 9/0 gunmetal seed beads
28g × 9/0 black seed beads
2 × 12mm mother of pearl domed buttons
8 × 2.5cm faceted beads, flat on one side, or domed buttons, oval shaped if possible
2 metres × 3mm black cord braid
EQUIPMENT
Thread
Beading needles
Embroidery hoop
Bamboo point turner
Sewing machine
Pinking shears
Dress stand
Making order
FOUNDATION BODICE
The original bodice has two layers of insertion lace hand sewn below the neck edge; this has not been included on the reproduction garment or on the pattern. There is a further trim of narrow lace around the armholes, which has also been omitted. To make the bodice, begin by inserting the gussets into the front panels: on the inside, with right sides facing, pin the gusset in place and machine. Before any further seams are machined the bodice should be fitted to the wearer to ensure a snug fit. To do this all seams should be pinned and tacked and then a fitting can take place. The original bodice has seams finished by oversewing but the reproduction can either be overlocked or the edges of the seam allowance turned over and machined with a straight stitch.
Centre back boning channel.
Inside view of the reproduction foundation bodice.
To form the boning channels, pin cotton tape on top of the open seams and machine down either side and across the top. Insert plastic boning into the channels and, using a zipper foot, machine across the bottom to secure the bone in place. The boning needs to be either melted at either end or bound with cotton tape to make sure it does not poke through the boning channels. The bodice fastens with hook and eye tape that is folded in half lengthways and machined on to the facing at the front of the bodice. Make the facings into boning channels and insert further boning at the centre fronts to the side of the hook and eye tape. Secure in place at the top and bottom. To finish the raw edges around the top of the foundation bodice a facing cut on the bias is sewn along the inside. This forms a casing which a cotton tape passes through in order to gather the neckline in to fit the wearer at the centre back. To make the facing, cut two bias strips 2.5cm × 50cm. Fold the bias strips in half lengthways and press and machine down the inside of one short edge on either piece. Place the strip along the outside of the bodice with the raw edges lining up. Machine along the seam line and trim and layer the seam to eliminate as much bulk as possible. Fold the facings to the inside of the bodice and in place. Leaving the centre back open, machine along the bottom of the casing. Attach a small safety pin to the cotton tape and pass through both casings. Secure in place at the centre fronts and leave the long ends of the tape at the centre back. The bottom edge of the bodice is finished with bias binding.
Evening gown foundation bodice pattern A.
Evening gown pattern B.
Evening gown pattern C.
Evening gown pattern D.
Evening gown pattern E.
Evening gown beaded panel F.
Casing around the neck edge of the foundation bodice.
Hook and eye tape is sewn at the centre front.
UNDERSKIRT
The underskirt is a long A-line skirt made from a front and a back panel. The underskirt opens at the centre front with a false placket. The underskirt is made like a lining so all darts and seams are on the outside leaving the inside, which sits against the body, looking neat and tidy. Begin by closing the darts: pin and machine the darts from the wide end to the point. Dressmakers may prefer to knot the ends of the dart rather than reverse stitching; as a costume maker time is often in short supply and so careful reverse stitching exactly along the same line is preferable. Press the darts inwards to face the centre front and centre back.
To make the placket, first apply the fusible interfacing to the reverse side and make sure all seam lines are still visible. The placket is then placed against the inside of the skirt, and taking a 0.5cm seam allowance from the newly cut opening, pin the placket in place with right sides facing and raw edges aligned. Machine down one edge until reaching the midpoint, lift the foot and pivot the skirt to get across the corner, machine up the other side, and press the seam allowance into the placket. Fold the placket along the fold line and turn the seam allowance under. Pin the placket in place and topstitch close to the edge.
Placket detail, centre front underskirt.
The foundation bodice with sleeves sewn around the armholes and the underskirt sewn on top.
The next process is to join the side seams: with right sides facing pin and machine along the seam lines. The seams on the original skirt are left unfinished but on the reproduction skirt they can be overlocked together before pressing towards the back. Press the hem upwards, fold twice to neaten the edge and topstitch in place.
The underskirt has a false front made from satin that sits below the placket and is shaped at the sides. To make the false front begin by hemming it, press over 1cm along the bottom raw edge and machine in place. With right sides facing press the hem upwards and pin down the sides. Machine the sides of the hem, snip off the corners, trim the seam and turn through. Press the hem lightly along the folded edge only and slipstitch the hem to the fabric taking care that the stitches do not show through to the front. Press the remaining seam allowances under and position the false front on top of the underskirt. It is useful to insert something to prevent pins from going through both layers, a Pattern Master is ideal. Pin the false front in place and then topstitch to the front layer by machining around the folded edge.
The underskirt is attached to the foundation bodice around the waist and sits on the outside of the foundation bodice. Prepare the underskirt for attachment by turning the seam allowance over inwards along the top edge and machining; this edge will be completely enclosed when the gown is finished. Place the bodice on the stand and pin the underskirt in place following the line indicated on the pattern. On the original bodice the underskirt is attached with tacking stitches by hand but it can be machined in place on the reproduction garment.
OUTER BODICE
The satin outer bodice wraps over at the front and back. The front bodice dips at the centre front and the fullness is reduced at the bottom edge with tucks and gathering. Begin by joining the shoulder seams: pin right sides together and machine along the seam line. Press the seam open; the edges of the original bodice are left raw but on the reproduction they are cut with pinking shears. The seams could be overlocked but there is a risk that the overlocking will leave an imprint on the right side of the satin when the seams are pressed. The shoulder seams have a pleat and this can be folded and machined in place on the reverse side.
Pleat sewn on the reverse side of the shoulder.
Before joining the side seams, finish the armholes by cutting two 4cm × 50cm bias strips. Fold each strip in half lengthways with the right sides facing outwards and press. Use tailor's chalk to draw a 0.5cm seam allowance in from the raw edge. Pin on the outside of the armhole with the raw edges facing the armhole. Machine in place and trim the armhole seams down to 0.5cm. Press the binding over the seams and to the inside of the armhole. Pin and tack in place and then machine in the ditch created by the seam.
The side seams can be joined by the same method as the shoulder seams. Large flat beads or buttons can be sewn on top of the shoulders at this stage. The neck edges of the bodice are finished with a fold of chiffon at either side. The pattern is shaped to a wider curve at the front and is straight at the back. Fold the chiffon in half lengthways along the fold and tack along the seam line. With right sides facing place the chiffon onto the bodice so that the raw edges are aligned. Pin and machine in place. Press the seams away from the neck edge and topstitch to hold the chiffon in place. This topstitching can then form the guide for a row of beads to be sewn on top with a running stitch. Take care not to sew beads where the bodice overlaps at the front and back.
Sew beads along the edge of the chiffon band.
SKIRT
Tack down the centre front lines to transfer the marking to the right side of the skirt and tack around the hem and along the top edge for the same purpose. Join the train to either panel by pinning right sides together and machining along the seam lines. Trim the seams with pinking shears and press the seams open. Machine two rows of gathering across the top of the train panel and pull to gather to reduce the top to 3cm. There are two small pleats marked on each skirt panel along the top edge and these can be folded in place and machined across the top, within the seam allowance, to hold in place.
The curved hem of the original skirt is finished by a 7cm bias cut facing. Cut the bias strips to include seam allowance, and press the raw edges in. To help form the curved edges of the hem sew a single row of temporary gathering stich at each curve within the seam allowance of the skirt, pull gently to gather and the seam allowance will begin to curl over; press along the folded edge. With right sides facing pin the facing in place and machine to the dress. The facing is then slipstitched to the back of the satin. Wide black bias binding has been used on the reproduction gown. The facing is sewn to the skirt with a long slipstitch that picks up one thread of the skirt at a time. Press the hem lightly along the edge when sewn to prevent the facing from showing on the right side of the satin.
The chiffon side panels hanging down from the waist are finished with a hand-rolled hem on the original gown. The thread used was not colour fast and so it is now a shade of brown. A rolled-hem foot was attached to the sewing machine to create a narrow hem on the reproduction panels. Pressing the first 10cm over twice makes the process of feeding the slippery chiffon fabric into the rolled-hem foot a little easier. When the panels are completed, using a long machine stitch, sew two rows of gathering stitches within the seam allowance. Pull the two top threads, gather until the panels measure 20cm and then knot the threads at either end.
SCARF
To make the narrow neck scarf begin by sandwiching the shorter panel between the two longer panels with a 0.5cm seam and press open. With right sides facing fold the band in half lengthways, match the seams and pin along the raw edges leaving an opening in the centre of 8cm for turning through. Machine around the edge and trim to 0.5cm, snip off the corners and use a long-handled wooden spoon, or similar, and turn the scarf the right way round. Use a bamboo point turner to make sure the corners are sharp, and press. Close the 8cm gap by pressing the seam allowance inwards, pin and slipstitch by hand. To make the decorative bobbles that hang from the four corners, cut four circles of satin big enough to cover each pompom (I used a large reel of thread as a template). Use a double thread and small stitches to gather each bobble over the pompoms. Join the bobbles to the corners of the scarf with a loose chain stitch 1.5cm long made from three strands of embroidery thread. Place a drop of Fray Check over the raw edges.
With right sides facing, pin around the edge of the scarf, leaving a gap in the centre for turning through.
Make the scarf bobbles from pompoms and circles of satin.
BEADED BODICE PANELS
Begin by layering the tulle. The original gown has four layers of lighter weight cream tulle, a back layer of stiffer tulle and a top layer of black tulle. I tacked two layers of cream tulle under the outer layer of black together to prevent movement. The other layers were tacked behind once the panel was completed. Place the template behind the net and pin at the corners. Trace the outline of each shape either with an air erasable marker or by tracing with thread. The flower motifs were constructed separately on a piece of black net. The flowers are cut out and then applied to the panel once all other sections are completed.
Beginning with the front panel, place it in an embroidery hoop and adjust so the tension is firm. To sew beads use a beading needle to thread beads onto and then use a small needle to couch the beads in place. Sew the sequins flat using a backstitch. For the swirls around the flowers, thread two beads and one sequin onto a needle and backstitch behind each sequin before repeating. The tension is quite loose for the swirls. The cord is lightly stitched in place using a stab stitch and is couched at the corners. The centre back panel has a smaller opening, so it is not necessary to complete another whole panel, just the section that will show. On the original bodice this is 11cm from the edge of the panel to the top of the chiffon. A folded piece of chiffon has been sewn across the bottom of the beading on the original panel.
Place the tulle in an embroidery hoop and transfer the lines from the template to the tulle.
Cut a length of chiffon to match the width of the bottom of the beading panel and 5cm wide. Fold in half lengthways. Place the fold along the bottom of the beaded section, pin and slipstitch to the tulle. Finish the long raw edge by folding the edges under and slipstitching to the tulle. To finish the tulle panel trim the seam allowance down to 1cm, fold it behind the panel and slipstitch to the tulle.
Cut around the sequin flower motif with small sharp scissors.
The completed front bib panel.
BLACK BEADED FLOWER MOTIF ON THE SKIRT
Cut out fifteen pairs of petals, adding seam allowance, and machine around the edge, snip off the point, place a dab of Fray Check at the point and turn the right way round (a chopstick is useful for this process). Make sure the leaf shape is smooth with the seams sitting along the edges, and then press each leaf. Pleat the bottom edge of each leaf into three. Take eight leaves and machine the pleats in place in a continuous strip. Close the circle of leaves by knotting the ends of the thread. Cut two circles of black net that are bigger than the circumference of the hole in the centre and slipstitch the leaves to the net. Place a further leaf in the space behind each leaf. Sew a bead at the point of each leaf with a single stitch of double thread, then wind the thread around the base of the bead twice before backstitching at the back of the leaf. Further beads are sewn in random places between the tips of the leaves and the centre. Sew a domed button in the centre of the flower and a row of beads around the button. To do this, thread a line of beads onto a beading needle, enough to sit around the edge of the button, and couch in place with a separate needle and thread.
Feed the ends of the petals through the machine and join in a strip before forming in a circle.
Layer the petals on top of a circle of black net.
BELT
The belt is comprised of four shaped beaded sections, for which a pattern has been provided. They sit at the centre front, centre back and at the sides of the waist. There are rectangular sections of beaded net in between the motifs measuring 3.5cm high and 8cm wide. On the original belt, the front motif has a circle of faceted beads in the centre; the three other motifs have a single flat bead sewn in the centre. The beading is sewn onto separate black net sections before being mounted, and then all sections are hand sewn to a straight satin belt. The reproduction belt has just one beaded motif sewn at the centre front.
To make the beaded motif for the centre front, place the motif template behind a larger rectangle of net and trace the outline onto the net with a white marker pen. The net can then be placed in an embroidery hoop if desired. I found it easier to work on a smaller piece of net which was the motif with a 2cm seam allowance around the edge; the disadvantage with this method is that when beading the cut edges of the net catch the thread.
To make the inner circle of beads, thread a beading needle with a single strand of thread and make a knot in the long end. Then at the edge of the inner circle bring the thread through to the front of the motif and backstitch. Thread eighteen beads onto the beading needle and starting at the outer edge of the circle wind the beads into a spiral and couch in place with a separate needle and thread. Then make the two surrounding circles by threading six black seed beads at a time onto a beading needle and couching every three beads. Two rectangles of bugle beads radiate outwards at either side, edged with a further two rows of seed beads. Bugle beads are sewn all around the motif at this stage before being edged with a final row of seed beads. Six faceted beads are sewn along the top of the motif. To make the beaded fringe that hangs from the lower edge of the motif, take a length of narrow bias binding 10cm long and fold in half. The beaded fringe is sewn to the folded edge and the completed section is then sewn behind the motif. Thread a beading needle with a single thread and tie a knot in one end and attach to the centre of the bias strip. Thread nineteen seed beads onto a beading needle and push onto the thread, add a further single bead and then push the needle back through the previous nineteen beads only and backstitch at the top. The other fringes are the same length and are spaced 2mm apart. When eleven columns have been made curve the bias tape and sew to the back of the motif so that the fringe hangs down at the front.
Working from the centre outwards, sew beads to black net to make the motif for the centre front of the belt.
To make the satin belt cut a rectangle of fabric long enough to fit around the waist and 9cm wide, which includes 1cm seam allowance. With right sides facing fold the band in half lengthways and pin along the raw edges leaving an opening in the centre of 8cm for turning through. Machine around the edge and trim to 0.5cm, snip off the corners and use a long-handled wooden spoon, or similar, and turn the scarf the right way round. Use a bamboo point turner to make sure the corners are sharp, and press. Close the 8cm gap by pressing the seam allowance inwards, pin and slipstitch by hand. Mark the centre front and centre back and the position of the four motifs. Pin the motifs in place and slipstitch to the belt. The belt is hand stitched to the dress once all sections are in place. The belt fastens at the left-hand side with a triangle comprised of two snap fasteners at the outer edge and a hook and bar sewn behind.
SLEEVES
Begin by hemming the long edges and then inner edges of the point. This is a rolled hem done by hand on the original garment and a rolled hem sewn by machine on the reproduction. To close the seam nearest the body a French seam is used. To make the French seam place wrong sides together and pin and machine along the seam line. Trim to 3mm and press the seam to one side. Turn to the right side and fold in half and press along the seam, encasing the raw edges. Sew a 4mm seam and press to one side. Attach the sleeves to the armhole of the foundation bodice. The sleeve is closed with a backstitch at 22cm down from the fold. Sew the beaded tassels to the four points at the bottom of the sleeves. Pin the sleeves to the armholes of the bodice and machine around the seam line; finish the edges by cutting with pinking shears. The seam is pushed into the bodice.
ASSEMBLING THE GOWN
Put the foundation bodice and attached underskirt on the stand, fasten and gather the bodice to fit by adjusting the cotton tape at the centre back. The line tacked around the lower section of the bodice is the guide for adding all parts of the body of the gown. Place the chiffon layers at the sides of the skirt and pin in place. Join the over bodice to the skirt at the waist and put on the stand with the right side wrapping over the left, over the foundation bodice. The flower motif is sewn to the dress from behind using a double thread and large stitches to enable it to be removed when the dress is in storage. The flower motif is positioned at the bottom of the overlap of the skirt when the dress is on the stand. The skirt is joined to the false front with two swing catches, each 2.5cm long, at the hem. Finally add the belt, hooks and eyes, and snap fasteners.
Assemble the evening gown on the stand.
ADAPTING THE EVENING GOWN
If the gown is to be worn repeatedly for a stage play the foundation bodice could be made from a stronger fabric such as coutil. The beaded belt could be replaced by a satin belt that fastens with a decorative buckle. The dress has many layers that could be left off or altered. The centre front and centre back tulle panels could be replaced with black lace backed with white silk. The sleeves and chiffon skirt panels could be removed and the foundation skirt lengthened to make a version of the dress similar to the illustration by Ida Pritchard used at the start of this chapter.
Chapter 10
Lined Cape | 10
---|---
_'The cutting of capes is a very simple matter. They must be properly balanced to secure the fullness falling in folds that will hang gracefully', stated_ Edwardian Ladies' Tailoring _in 1910. In the same period The Lady's World showed designs for three short summer capes with the recommended fabrics being silk, velvet or chiffon. It was suggested each cape could be mounted 'upon a plain lining of silk'. The yellow cape featured in this chapter is from Royal Pavilion & Museums, Brighton & Hove, museum accession number C003367.3. It is a half-circular cape made from a silk grosgrain fabric with layered organza and chiffon ruffles around the front edge and along the seam between the cape and the flounce that passes around the hem of the cape. It was purchased by Katherine Farebrother from Dickins and Jones, a smart London department store, in c.1900 along with a coordinating long gown. Although the cape has the appearance of being a decorative and frivolous garment, it is in fact also a practical one. Inside it has a quilted, padded lining for warmth, and a roomy and useful patch pocket._
The reproduction cape.
An illustration of the front of the finished cape. (Royal Pavilion & Museums, Brighton & Hove)
Detail of the inside of the original cape showing the patch pocket.
It fastens edge to edge at the centre front with three sets of hooks and eyes. The collar is rounded at the front edge and has ruffles on the top and bottom, which suggests that it could be worn either flat to the body or pulled upwards and worn as a high collar. Edwardian fashion journalists such as Mrs Pritchard, writing in _The Ladies' Field_ , stressed the importance of clothing that was 'becoming' to the complexion, the face and the figure, suggesting that this collar could have been used to softly frame the face of the wearer. _Weldon's Practical Needlework_ wrote in more specific terms about the advantages of frilling around a collar, stating: 'it gives the look of fluffiness and fullness that a thin neck needs.' Enlarge the pattern, add seam allowance and use it to cut and make a toile to test the shape and fit of the cape before cutting in fabric.
MATERIALS AND EQUIPMENT
3.5 metres yellow silk grosgrain fabric, or similar
2 metres yellow chiffon
2 metres white chiffon
Three sets of size 3 silver hooks and eyes
1 metre lightweight wadding or batting
1 metre domette interlining fabric
3.5 metres ivory habutai lining
20cm collar canvas
Thread
Sewing machine
Bamboo point turner
Rouleau loop turner (optional)
The original cape with the collar turned up.
Cape pattern.
Cutting
Place the pattern on the fabric; by following the grain line indicated on the pattern, the centre back will be cut on the bias. Use tailor's chalk to draw around the outline of the pattern and then transfer the shape to the other side of the fabric by placing carbon paper underneath and wheeling around the edge with a tracing wheel. Add a generous seam allowance. Transfer pattern markings for the darts and other balance points to the fabric using tailor's tacks. Use the upper cape pattern to cut two pieces of wadding but do not add seam allowance. Cut into the darts and remove the V shape of each dart; the wadding sits edge to edge rather than darts being sewn in the bulky wadding. Cut canvas for the under collar and add seam allowance.
The flounce is cut in sections with the centre back sitting on the bias. The lining is cut from the same pattern pieces as the cape and needs seam allowance; a single layer of domette should also be cut from the flounce pattern pieces. The top and bottom collar pieces are cut from the same fabric and also need seam allowance. The bottom collar pieces also need collar canvas cut using the pattern pieces, with added seam allowance. Use carbon paper and a tracing wheel to transfer markings to all pieces of the collar canvas.
The ruffles are cut from strips measuring the width of the fabric, which is 150cm. The yellow strips are 12cm deep and the upper, white strips are 10cm deep, which includes 0.5cm seam allowance. A total of five ruffles are needed to sew around the cape and a further length is needed if a ruffle is required for the under collar. Further strips are needed for the upper collar. A yellow ruffle sits around the outer edge, which should be cut 16cm deep, to be folded to 8cm. A small white ruffle that sits near the neck measures 10cm in total and 5cm when folded. Cut the pocket and add seam allowance; cut a further pocket lining section from the same fabric.
Cut the collar canvas and add seam allowance.
Making order
The outer cape
Separate all pieces and begin by joining the centre back seam. Place right sides together and pin and machine along the seam line. Press the seam open. To form the darts, working on the wrong side of the fabric fold the dart down the centre, pin in place and machine from the wide end to the point. Cut away excess fabric to make open darts and press flat. To prevent the neck from stretching machine a row of stay stitching using a small machine stitch, within the seam allowance. To prevent the centre front sections from stretching machine a narrow cotton tape on the reverse side just to the inside of the seam allowance.
Form the darts in the cape, cut and press open.
Sew narrow tape to the inside of the seam allowance.
Flounce
Mount the lining sections of the flounce by placing a layer of domette on the reverse side of each piece. Use large basting stitches to hold the domette to the lining, making sure that the layers are lying flat. This will help to give a smooth finish. Seam the lining sections together and press the seams to one side. Seam the outer flounce sections together and press the seams open. Working on a large flat surface, lay the flounce with the right side facing upwards and lay the lining on top so that the right sides are facing. Lay all seams on top of each other and pin in place around the bottom edge only, along the seam line. Machine around the edge, trim the seam allowance down to 1cm and snip into the seam allowance at intervals to give the flounce a smooth curved edge. Turn the right way round and press. Lay the finished flounce back on the flat surface and pin and tack the top raw edge together, smoothing the lining upwards as you go.
Attach the flounce to the bottom of the cape with right sides facing and raw edges aligned. Line up the centre back seam and push a long pin through both seams – this can be kept in place while sewing to make sure the seams match perfectly. Pin and tack the flounce in place and machine along the seam line. Press the seam towards the inside of the cape.
Pin and machine the flounce to the edge of the cape.
Collar
To make the collar, begin by laying the collar canvas on the reverse side of the under collar pieces and then secure in place by tacking along the seam lines. With right sides facing pin all under collar pieces together along the seam lines and machine. Trim the collar canvas by cutting close to the line of stitching. Trim the collar seams and snip at intervals to allow the seams to sit smoothly once pressed. Press the collar over a tailor's ham or similar. Pin and machine the upper collar sections together, trim the seam allowances and press the seams open.
To join the upper and under collar place right sides together and pin and machine around the top edge; the bottom edge is left free for joining to the cape. Layer the seam allowance to avoid a bulky seam and snip at intervals. Turn the right way round and press along the edge (use the bamboo point turner to push the seam outwards). Check that both edges of the collar are the same size by folding it in half, and adjust if necessary. To finish, sew a temporary row of tacking around the top edge of the collar.
Ruffles
The ruffles edging the original cape are self-faced double ruffles made from long strips folded into three and gathered in the centre. On the original cape they are made by hand and joined to the cape with a small running stitch down the centre of each ruffle. The reproduction ruffles are made by folding each strip in half lengthways and machining a 0.5cm seam. A rouleau loop turner is useful for turning the ruffles the right way round or a safety pin can be attached to one end and pushed through. To form the ruffles pin one yellow strip to a firm surface such as an ironing board or cork board with the seam lying in the centre and underneath. Place a white strip on top and centre it so that an equal band of yellow is visible along either edge. Take a long needle, threaded with white double thread, tie a knot and sew a backstitch and then sew a running stitch down the centre with stitches approximately 1cm long. When reaching the end pull the thread to gather the ruffle strip, distributing the gathers evenly while pulling until the ruffle measures 60cm. Fasten the thread with a backstitch. The ruffles are attached to the cape by hand because a machine stitch would make them look too stiff. To attach the ruffles to the cape, place the cape on a stand and pin the ruffles in place around the cape following the edge of the upper cape. Place pins horizontally and turn the raw edges of the ruffles under as you go. Once all ruffles are in place sew in place with a double thread and long back stitches.
The length of the frill can be accurately checked by pinning it to a cork board alongside a tape measure.
The lower collar ruffles can be added before the collar is attached to the cape. The upper collar ruffles should be pinned in place with the cape on the stand once the collar has been securely sewn in place. The collar has the same ruffles as the edge of the cape sewn on the under collar, this should be done before the top collar has its ruffles attached.
The top collar has a slightly different type of ruffle. Cut a width of white chiffon 32mm deep and fold in half lengthways and sew a 0.5cm seam. Turn the right way round and, with the seam lying in the centre and underneath, place pins across the width of the strip. Take a long needle, threaded with white double thread, tie a knot and sew a backstitch and then sew a running stitch down the centre with stitches approximately 1cm long. When you reach the end, fold the strip in half lengthways and pull the thread to gather, distributing the gathers evenly while pulling until the ruffle will sit along the top edge of the collar.
Cut a final fabric width of yellow chiffon and fold in half lengthways and machine a 0.5cm seam, then turn the right way round. Flatten the strip so that the seam runs along one edge. With the long needle threaded with white double thread, tie a knot and sew a backstitch and then sew a running stitch down the edge of the seam with stitches approximately 1cm long. Pull to gather. The top collar ruffles are pinned in place so that the gathered edges are resting up against each other. Slipstitch in place.
Quilted lining
To make the quilted lining, first mark the lines for quilting on the right side of the lining. Use an air erasable marker to mark the quilting lines, which are 6cm apart and run diagonally at right angles to the centre back. The lining sections are quilted separately before being joined together. To make up the quilted lining place the layer of lightweight wadding on a flat surface and place the lining piece on top; baste together with large stitches. Machine the quilting lines making sure that each line is sewn in the same direction to prevent the fabric from puckering. The lining may have moved during the quilting process so place the pattern piece back on top of the quilted pieces and re-mark if necessary. Join the centre back seam, taking care to match the chevrons created by the quilting. Tack the wadding to the fabric around the darts in the lining and pin and machine the darts in place, sewing from the wide end to the point. Press flat from the right side of the cape. Before attaching the lining to the cape, fold the centre front openings to the inside along the seam line and press along the folded edge. Tack and herringbone lightly to the cape taking care not to let the stitches show on the right side.
Mark chevrons on top of the lining using an air erasable marker and Pattern Master.
The hooks and eyes are sandwiched between the cape and the lining and so need to be sewn on before the lining is attached. They are spaced 6cm apart with the first set of hooks and eyes starting 1.5cm down from the neck seam line. Sew the hooks and eyes in place, beginning with a hook on the right hand side and alternating with a bar before a final hook. Sew corresponding fastenings on the other side.
Patch pocket
To make the patch pocket, first mount each section with domette, which will help to give a smooth finish. Then with right sides facing join the pocket lining to the pocket along the straight edge leaving a gap of 6cm for turning through. Press the seam and press lightly along the fold that forms the top of the pocket. With right sides facing, pin around the edge of the pocket and machine along the seam line. Snip off the top corners and clip around the curved edge at 1cm intervals. Turn the pocket the right way round by pulling through the opening. Use a bamboo point turner to push the corners out and to make sure the edge is crisp. Press lightly and then close the gap by slipstitching by hand. The pocket on the original cape is not quilted but has 3cm wide diagonal lines of machining following in the same direction as on the cape. Place the pocket on top of the right front cape lining and position it 17cm in from the front edge with the bottom edge resting just above the seam line. On the museum's cape the pocket is slipstitched around the edge; it can, however, be topstitched to the lining by machine on the reproduction cape.
Finishing
To add the lining to the cape, lie the cape on a flat surface with the wrong side facing upwards. Lay the lining on top with the right side facing upwards and line up the centre back seam and darts, pin in place all around the edge and through the centre back seam. Fold the edge of the lining under and pin and tack to the edge of the cape. Smooth the lining upwards towards the neck edge and pin along the seam allowance at the neck edge. To test that the lining is not pulling, place the cape on a dress stand or on a person. If the cape is hanging smoothly then slipstitch the lining to the cape around the edge.
Add a hanging loop to the centre back neck.
There is a hanging loop at the centre back neck made from a rouleau loop. To make the loop, cut a strip of lining fabric on the bias 3cm × 20cm. Fold in half lengthways and machine 0.5cm away from the fold, pulling the bias strip as you sew. Trim the seam allowance down to 0.2mm. To turn the right way round, insert the rouleau loop turner, catch the top and pull. Press before forming into a loop measuring 9cm. Place the raw edges into the centre back seam allowance and machine the loop in place.
Machine the under collar to the outer neck edge and slipstitch the upper collar to the lining.
Place the cape on a stand and pin the frill around the edge of the upper collar.
To attach the collar match up the centre back marks and with right sides facing pin the under collar around the edge of the neck, keeping the top collar out of the way. Tack in place and then machine along the seam line with the collar on the top as you sew. Trim and grade the seam allowance and then snip into the seam allowance. Press the seams into the collar over a tailor's ham or a sleeve board and fold the bottom raw edge of the top collar under and pin and hand sew to the lining using a small felling stitch. Add the large white collar ruffle around the edge of the upper collar and sew in place with a small running stitch. Finally add the narrower single yellow ruffle to slightly overlap and sew in place with a small running stitch.
ADAPTING THE CAPE
To adapt the cape for daywear a wool or serge fabric could be used and the ruffles could be omitted. The quilted lining can be removed and a cotton mounting fabric used instead. The cape could also be made in black satin with black lace ruffles or guipure motifs, and black silk cord ties for glamorous evening wear. To adapt the cape for a quick change just one hook and bar can be used.
Chapter 11
Evening Bag, Hat and Parasol | 11
---|---
_An Edwardian ensemble was incomplete without the addition of a range of accessories and many possible candidates in the museums' archives were considered for reproduction in this chapter. The three projects were selected because it was possible to source similar fabrics and trimmings and because it was possible to adapt them to a range of other styles._
_The small drawstring bag was a useful yet elegant addition to evening wear for an Edwardian woman. Described in advertising and editorial features as either a 'Dorothy bag' or by the French term réticule, they were made from silk or velvet and were either encrusted with beads, sequins or tassels, or sometimes all three. A crocheted or macramé version was also available for day use and craft manuals and journals featured many examples for readers to make. Worthing Museum has a small purple velvet Dorothy bag adorned with a floral trellis pattern around the outside with a central floral motif, formed from Berlin steel beadwork. Dorothy bags were available to purchase at department stores but dressmakers could also make their own versions._
Millinery was a highly specialized occupation and Edwardian women had a wide choice of skilled milliners to visit when a new hat was needed. They also had the option of making their own hat by following patterns and instructions that were supplied in dressmaking manuals and journals. The parasol featured in this section is quite fancy but for a woman on a lower income an unlined parasol covered in a striped canvas fabric was a fashionable accessory and a practical sunshade for a walk in the park on Sundays.
The reproduction accessory projects.
Postcard of a woman carrying a striped parasol and wearing a walking suit and feather boa. (Private collection)
DRAWSTRING BEADED EVENING BAG
The bag featured in this chapter comes from Royal Pavilion & Museums, Brighton & Hove, Leeson and Vokins sub-collection, museum accession number CTTMP000277. Although the bag is undated it is a style that can be seen in journals and dressmaking manuals throughout the Edwardian period. The original bag is made from black grosgrain satin with a dull sheen and is fully lined with black silk. There is a densely beaded exterior section on both sides of the bag comprised of flowers, hearts and abstract swirls. The top edge is scalloped and edged with two rows of black beads. A casing on the outside is threaded with silk cord, which is finished off at either end by round beaded bobbles. Due to the weight of the glass beads the bag feels heavy but it is also robust and in excellent condition. The reproduction bag is beaded on one side only and features a simplified version of the beading pattern designed and beaded by Rachel Woolcott.
The original beaded evening bag from Leeson and Vokins department store, Brighton. (Royal Pavilion & Museums)
Detail of the bottom of the original beaded evening bag.
MATERIALS AND EQUIPMENT
25cm Duchess silk
25cm lining
25g × 8/0 silver lined chocolate rocaille beads for the main swirl and hearts
25g × 7mm bugles col. 1870
Gütermann to fill the flowers
25g seed beads col.1480
Gütermann to outline the flowers
25g × 8/0 black rocaille beads for edging
40 × 0.5cm bright gold balls
2.2 metres × 3mm black lacing cord
Embroidery hoop
Beading needles
Thread
Fray Check
Evening bag pattern.
Cutting
Copy the pattern from the book and cut around the edge. Do not add seam allowance at this stage. Place the fabric face down on a flat surface, layer carbon paper face down on top and finally the template facing upwards. Use a small tracing wheel to transfer the pattern onto the reverse side of the bag. Thread mark to transfer the marking to the front of the bag. Cut the lining using the same pattern, and add 1.5cm seam allowance around the outside.
Making order
Each bead on the reproduction bag was sewn on individually with a running stitch. They could also be sewn by the couching method, which involves threading a few beads at a time onto the beading needle and using a second short needle and thread to catch the beads and thread down, either behind every bead or every few beads. To begin beading the bag, place the marked bag section in an embroidery hoop with the right side of the fabric facing upwards. Using a beading needle and single thread start with a knot and backstitch on the reverse side of the fabric and start beading the inside of the flower shapes with bugle beads. Fill the flower with two bugle beads in each petal and complete the flowers by outlining with a further shape of bugle beads. Form the long swirling shapes and heart shapes by sewing three long rows of seed beads next to each other. Finally, randomly scatter and individually sew on the bright gold balls. Repeat the beading pattern on the other bag piece if desired.
Once the beading is complete the next stage is to add the casings on the outside of the bag for the cord to pass through. The rectangular casings are cut on the straight grain and measure 20cm × 3.5cm. Press all edges of the casings under by 1cm. With right sides facing place the seam line of the casing piece along the top row of beads and pin in place. Attach a zipper foot to the machine and machine along the seam line. Making sure the raw edges are still folded inwards, sew the top edge of the casing in place and hand sew the top of the casing using a small needle and felling stitch. Repeat on the other bag piece. The next stage is to sew a row of small black beads along the top of the casings. Thread three beads onto the beading needle and make a backstitch to form a small pyramid effect. Leave a space of 13mm between each pyramid.
Transfer the markings to the wrong side of the fabric using a tracing wheel and carbon paper.
Pin the casing to the front of the bag.
Sew beads along the top of the casing.
Before joining both bag pieces together trim the excess fabric away to leave a 1.5cm seam allowance around the bag. With right sides facing place the bag pieces together and pin through the seam lines. Still using the zipper foot, machine carefully around the edge with the needle positioned to the left, lifting the foot and pivoting when necessary to prevent the fabric from dragging. Trim the seams and snip at intervals. Join the lining seams at the sides, leaving the bottom curved edge open by 5cm for turning the bag out. With right sides facing position the lining over the bag and match up the raw edges along the top and the side seams.
With right sides facing slot the bag into the lining.
Slipstitch the bottom of the lining.
Pin and machine along the scalloped edge, using the zipper foot; leave the needle in the fabric and pivot the bag when you reach a dip in the scallops. Trim the scalloped edge down to 0.5cm and snip into the curves and into the points of the scallop. Dab Fray Check onto the scallop points and allow to dry. Turn the right way round by pulling through the gap in the bottom of the lining. Finish the beading by sewing two rows of black seed beads down the sides at the front of the side seam to form a continuous row all around the edge of the bag. To close the lining gap slipstitch by hand.
A long cord strap is made from a piece of cord measuring 110cm. Fold the cord in half and tie a knot in the end leaving a loop measuring 4cm. Place the raw ends inside the bag and pin in place at the side seams 5cm down from the top edge of the bag, and oversew to secure. Thread a single cord measuring 55cm through each casing and join the cords together by forming a knot at either end. Trim the frayed ends and hand sew over the edges. The knots on the original evening bag are covered in small black beads and this process can be done at this stage if required. To close the bag hold the knots at either end of the cord threaded through the casing and pull the cords.
ADAPTING THE BAG
The Dorothy bag can be made to coordinate with an outfit by using the same fabric. A variety of floral patterns can be sewn to the outside either with embroidery stitches or by beading. Collector Marion May has a black silk drawstring bag decorated on both sides with small fuchsias and roses made from coloured net.
The reproduction evening bag showing the position of the cord.
VELVET TOQUE
The velvet hat chosen for this chapter is from Worthing Museum and is described as a toque, which is a turban or beret style, mostly brimless hat. The museum accession number is 1966/860. The hat was made by R. Sayle & Co., Cambridge, and is dated c.1912. It is made from mushroom brown cotton velvet and is lined inside with black silk. The hat is constructed over a domed cage shaped with millinery wire. The crown is covered by ruched velvet and the brim is separately covered with pleated velvet. The decoration across the front of the brim is made from a band of black net decorated with a wavy pattern resembling autumn leaves. The leaf motifs are made from wool and gold thread with a further and larger wave of antique gold lace winding between them. A large dusky pink rose adorns the left side of the hat. The rose is made from velvet and organza with pale yellow tipped stamens with dark green velvet leaves placed behind.
MATERIALS AND EQUIPMENT
Decorative braid
75cm × 25cm black net
1 ball of Wendy Roam Fusion 4ply wool
Fine gold thread
1 metre × 2cm antique gold lace braid
Inks and a paintbrush for adding shading to the braid
Craft glue (optional)
Embroidery hoop
Hand sewing needles
Hat
1 metre lightweight brown cotton velvet
25cm black lining 50cm × 1cm black ribbon
3 metres × 1.2mm cotton-covered millinery wire
Strong thread for securing the millinery wire frame
Pliers
Sticky tape
Thread
Tailor's chalk
Sewing machine
A hat block (useful but not essential)
Silk rose
Front view of the original museum hat. (Worthing Museum and Art Gallery)
The reproduction hat.
Leaf motif template.
The reproduction hat is a close copy of the original hat although it has been made larger around the circumference of the head at 57cm; the original hat measures just 47cm. The silk rose would not have been produced by a dressmaker and therefore instructions have not been included for reproducing it, although details of the supplier are given at the end of the book.
Cutting
Cut a long strip of velvet on the bias to cover the brim, 125cm long and 18cm wide. Cut an oval of velvet on the bias, 68cm long and 56cm wide, to cover the crown. Cut a rectangle of lining 60cm × 18cm and a further circle approximately 7cm in diameter.
Making order
MILLINERY WIRE FRAME
Begin by straightening the millinery wire using thumbs and fingers; this will prevent the wire from distorting once made up into the frame. The millinery wire frame is formed of three oval shapes joined together by strengthening struts – the crown, the head size and the brim. The head size is the circumference of the head plus ease.
Make the millinery wire ovals using a template drawn on paper.
Make a pencil diagram on paper showing the dimensions of the measurement taken from around the head (head size), the crown and the brim. The head size measurement used for the reproduction hat is 57cm or 22½ inches, which is the average size for a hat. Position the head size in the centre and mark centre front and centre back. The brim is 7cm deep at the centre front and narrows to 4cm at the centre back. The crown measures slightly less than the head size at 52cm. Use these ellipses as a template to form the wire frame for the head size, crown and brim. Use the pliers to cut the millinery wire to the correct size.
Join each oval by overlapping the edges by 3cm. Cover the join with sticky tape and then wrap with strong thread or embroidery thread. To make the struts that join all three ovals together make a small L shape at the end of the millinery wire and place alongside the inside of the brim at the centre front. Measure 8cm and form another L, which is joined to the centre front of the head size. Then form the centre back strut, which measures 4cm, and add two further struts to fit at the sides. Form the crown as a cage with two wire arches at right angles (crossed at the crown) connecting the brim to the crown piece. The crown is 11cm high. Use thread to bind the joins between the wire hoops and the spokes of wire radiating outwards. Bind all the joins with sticky tape and then wrap securely with thread.
Make the cage for the hat.
The next process is to cover the frame with net. Cut the crown tip piece first, exactly to fit the oval of wire, and oversew the raw edge onto the wire. Cover the brim in the same way but leave a 1.5cm allowance around the inside of the head-size band and clip the curves to help this piece fit to the wire. This will fold upwards inside the hat to begin to form the band. Oversew the edge of the net onto the wire. Cut a rectangle of net long enough to lap the sides of the crown, approximately 59cm and as tall at the hat (11cm) and cover the sides. Use a 1.5cm wide strip of calico cut on the cross to bind the edge of the brim all the way round, stab stitching closer to the raw edge than to the wire. The binding should be taut and smooth and disguise all the joins between the bits of wire.
Cover the cage with net and bind the edge of the brim.
BRIM
To cover the brim, begin by running a long gathering stitch along both long edges of the velvet strip (alternating a longer stitch on the wrong side, short stitch on the right side of the fabric). Seam the short ends of the strip together to make a continuous loop, and mark halfway along the length of the strip with a pin. This will be the centre front of the brim, and the seam will be positioned at the back. Fold the loop in half lengthways with wrong sides facing. Using the gathering threads, ease the velvet band around the brim, ruching up the fabric and distributing the folds as evenly as possible. The brim will be sandwiched or enclosed with the fabric, with the fold along the outer edge of the brim and the raw edges at the innermost edge. Stab stitch the fabric into place on the wire frame, keeping stitches hidden among the folds of velvet as much as possible. As the brim gets narrower at the back of the hat, there will be excess fabric along the edge, which can be trimmed away.
CROWN
To form the crown take the large oval of velvet, cut on the bias, and mark the exact centre of the oval with tailor's chalk. Mark a smaller oval in the centre slightly larger than the dimensions of the oval used to form the crown of the wire frame. Taper the outer corners outwards slightly to make them more pointed (like the shape of an eye). Outline this shape with two parallel rows of gathering stitch 1.5cm apart. Use these threads to make a gently ruched shape at the centre of the fabric, and tie the ends of the threads in a knot to secure.
Add the brim cover.
Run a gathering stitch around the edge of the fabric oval all the way round, 1cm in from the raw edge. Using the threads from these stitches, gather the fabric to form a domed shape with a ruched section at the crown. Ease the fabric shape over the wire frame, positioning it with the one of the V shapes at either side of the crown section pointing towards the centre front. Distribute the folds of velvet as evenly as possible, and gather to fit snugly around. Stitch as inconspicuously as possible to the brim, over the edge of the velvet covering the brim, folding under and enclosing all raw edges.
Gather the crown by sewing double rows of running stitches in an ellipse.
Add the lining to the inside of the hat.
MAKING THE LINING
Along one long edge, fold and press a 0.5cm hem. Along the other long edge, fold, press and stitch a 1.5cm hem, which will be the casing for the ribbon drawstring. Fold in half widthways with right sides together, and pin the short sides together. You will now have a cylinder. Stitch this seam, leaving the end with the drawstring casing open, and thread the ribbon through the casing and draw the ribbons to gather the tip, and tie in a knot.
Take the 7cm lining circle and apply fusible interfacing to the reverse side. Glue or stitch this inside the crown of the hat at the apex of the crown. Position the lining inside the hat, with the seam at the centre back. Pin the lining fabric to the velvet at the inside edge of the brim around the head-fit band, easing in any fullness. Slipstitch into position. Adjust the drawstring if necessary. The hat is now ready for trimming.
DECORATIVE BRAID
To make the decorative braid that runs across the brim, take a single piece of black net. Press the net with a steam iron and a pressing cloth if needed. Thread mark a rectangle in the centre of the net measuring 70cm × 6cm, leaving a border around the edges. Trace off or print the leaf motif. Join the leaf motifs together ensuring that every other leaf faces upwards. Place the template behind the net and pin to an ironing board or cork board. Use the white marker pen to trace around each leaf shape. The leaf motif is formed by lightly couching two strands of wool around the outline of each leaf. This is a tricky process and it might be easier to use a craft glue to stick the wool in place before couching. To form the motifs, cut a long piece of wool, approximately 1 metre long and fold in half. Place two strands of wool over the outline and use a small needle and a single thread to stitch the wool to the net. Use a light tension. A fine gold thread is couched on top to finish the motifs. Pin the wider gold braid in a wave pattern between the leaves and hand sew in place with a small running stitch. When finished trim the braid down to the tacked line.
Make the hat braid by couching wool on net, following the shape of the leaf motif.
ADAPTING THE HAT
Women from across all classes of Edwardian society wore hats and this particular hat shape could be dressed up or down. A simpler version could be made without the braid trim and with the addition of a small bunch of violets at one side rather than the rose. A more elaborate version could be made in silk velvet with the addition of a plume of feathers and silk and velvet flowers. The hat brim could also be widened by constructing a larger brim from millinery wire. Writing in _The Ladies' Field_ , 3 February, 1906, the fashion journalist 'Penelope' suggested improving a toque with the addition of a 'soft diaphanous veil, draped around the shape and floating behind.'
FINISHING THE HAT
Pin the braid across the front of the brim and slipstitch in place. Oversew the flower in place using a double thread.
LACE-TRIMMED PARASOL
The parasol recreated in this section is based on a lace-trimmed parasol from the collection at Worthing Museum, museum accession number 2008/341. The museum has several fine examples of parasols wrapped in acid-free tissue paper and stored in large flat drawers in the archive. The examples at the museum are fully lined, or where the fabric has disintegrated it is possible to see where the lining was once attached. Edwardian parasols at the higher end of the market were often highly decorated with lace, ribbons, ruffles and frills and were often made to accessorize a specific outfit. One such example at Worthing Museum is a small black lace-covered parasol with a carved ivory stick, which is lined with mauve silk and finished with a grosgrain ribbon tied around the ferrule.
The reproduction parasol.
Small black lace parasol with mauve silk lining, mauve grosgrain ribbon tied around the ferrule and an ivory stick. (Worthing Museum and Art Gallery)
Detail of the original parasol handle with silk cord looped around the stick.
The end of the original parasol handle painted with a portrait of a young woman.
The parasol chosen for this chapter has a ceramic ball at the end of the handle featuring a handpainted portrait of a young woman wearing a Georgian style dress. This has not been attempted in the reproduction parasol. A cream silk decorative tassel is looped around the lower end of the long wooden stick and could easily be undone and used to wrap around the parasol and use as a tie when not in use. Ten layers of embroidered ivory cotton lace are layered down the parasol, each layer slightly overlapping the one below. The layers are sewn to the parasol with small running stitches. The lace is also caught down in vertical lines at approximately 6cm intervals, which would have kept the lace in place when the parasol was folded and held upside down by the handle. A wide taffeta ribbon is tied around the tip of the parasol, known as the ferrule. The parasol is fully lined in plain cotton fabric and so would have been effective as a sunshade.
Inside the original parasol showing the cotton lining.
The reproduction parasol has been made using a parasol kit, which is supplied with a basic frame, handle, stick and separate ferrule. A reproduction parasol could also be made by unpicking the cover from an existing parasol or umbrella and using the cover as a template.
MATERIALS AND EQUIPMENT
Adult parasol kit (see list of suppliers)
14m × 5cm lace
1 metre ivory cotton fabric (2 metres if lined)
Thread
50cm × 5.5cm wide satin or taffeta ribbon
PVA craft glue
1 readymade cord with tassels
Cutting
If using the parasol kit make a pattern by drawing an inverted T shape on pattern paper using a set square. The horizontal line measures 25cm and the vertical line is 42cm. At the tip of the vertical line square out by 2mm on either side then join the points to form a triangle. The vertical line forms the grain line. If using a different parasol frame, make a triangular pattern by measuring across the space between the spokes, find the midpoint and use a set square to square upwards. Make sure both sides of the triangle are the same shape. Add 1.5cm seam allowance around each side. Cut out enough triangles to complete the parasol; for the kit version ten triangles are needed. The museum's parasol is completely lined and so if making an exact copy then a further ten triangles are needed.
Making order
The reproduction parasol is not lined and therefore the seams are French seams. If a lining is added then the triangles can be joined with a single seam. The following instructions are given for an unlined parasol. To make a French seam place wrong sides together and machine a 1cm seam so that the raw edges show on the right side of the parasol. Trim the seam allowance away close to the line of stitching. Press the seam to one side and then fold along the seam with the raw edges enclosed inside. Pin and machine a 0.5cm seam. Press the French seams to one side. Before adding the lace, check that the parasol cover is a snug fit by pinning the cover to the frame. If the cover fits remove it from the frame and turn the bottom hem twice and machine a narrow seam. Lay the cover on a flat surface with the right side facing upwards. The first layer of lace hangs over the edge of the parasol. Take the lace and place the raw edge along the top of the seam and pin in place, adding a tiny pleat at the corner of each triangle if needed, as the lace winds around the parasol.
Make a pattern for the reproduction parasol.
Machine the lace onto the cover.
Hand sew the cover to the frame.
To pin the subsequent layers of lace in place, the cover needs to be placed back on the parasol. Pinning lace around a large parasol is a timeconsuming process and best tackled in short bursts. The lace on the original parasol would have been attached in this way, with each layer being sewn with a small running stitch. The lace on the reproduction parasol has been machined in place. When all layers are in place, using a needle and double thread sew a row of gathering stitches in the seam line at the top of the parasol cover and leave long ends for gathering. Tie a strong knot and firmly sew the tips to the ends of the French seams on the parasol cover using a double thread and backstitch a few times. Place the parasol cover over the frame and line up the seams with the ribs. Slip the ends of the ribs into the tips. Loosely sew the ribs to the French seams about halfway up the frame. Place the handle on the end of the stick and glue in place. Catch the ends of the scalloped edges to the cover with a backstitch where the lace meets the seams – this will hold the lace in place when the parasol is held upside down.
Secure the cover around the top of the stick by sewing a row of gathering stitched around the top of the cover.
Add additional lace and glue the ferrule in place with PVA.
Pull the gathering stitches at the top of the cover. Pull tight and wrap the thread around the ferrule before backstitching in place. Cover with a layer of gathered lace and one final layer with the scalloped edge facing upwards towards the top of the ferrule. Place the ferrule on the top of the stick and glue in place. Tie the ribbon around the ferrule and create a bow, each loop of the bow measuring 10cm in length. Finally, tie the tasselled silk cord around the handle.
ADAPTING THE PARASOL
Parasols came in a variety of sizes and some were decorative rather than practical. To reduce the making time for the parasol in this chapter a child's frame could be used. A simpler version of the parasol could be made in a cotton stripe without lace trimming. A parasol could also be made in a contrasting fabric to match the day dress project in Chapter 8. To make a more elaborate parasol based on an example from Worthing Museum a smaller frame could be used, which could then be lined in mauve silk and covered in a layer of black lace. A mauve ribbon can be used to tie in a bow.
Chapter 12
Wearing Edwardian Fashion | 12
---|---
_For all women, no matter what their financial circumstances might be, dressing with care and attention was seen as a sign of good manners. Women knew that their clothing and shoes should be clean, their belts worn taut and their hair brushed and dressed. Rules of etiquette dictated what was to be worn for each occasion and how a woman should behave. This was an age where a respectable woman would perch in an upright position on a seat rather than relax in comfort. Lady Colin Campbell explained in_ Etiquette of Good Society, _'In the house a woman should sit still and not fidget however a man may change his position in an infinity of ways, lounge and loll and cross his legs.' Strongly worded advice was directed towards the schoolgirl in_ Weldon's Practical Needlework:
It is in very bad form to go to an afternoon party in morning costume. A girl may regard herself as too insignificant a being for it to matter much what she wears...but she should dress from the point of view of her hostess's importance, not from her own humility.
A hand-painted sketch by Ida Pritchard of a woman wearing a blush pink gown with a lace overlay, and a large hat decorated with silk flowers. (Worthing Museum and Art Gallery)
Advertisement for the Frank Patent Skirt Band, a patented device for holding a blouse neatly inside a skirt. (© EMap and the London College of Fashion Archive)
Time spent thinking about what to wear was a preoccupation of the Edwardian woman of financial means. It was not merely a matter of getting dressed – clothes had to sit on the body to achieve the desired silhouette. Ideas came from journals, department stores, society pages and letters between correspondents. The matter also taxed the minds of inventors and there are several examples of patented devices that were meant to assist the fashionable woman in this matter. Keeping a blouse neatly tucked into a skirt and pouched at the front led to the invention of patented skirt bands. In her memoir _Testament of Youth_ Vera Brittain recalls a contraption worn during her Edwardian school days:
We wore green flannel blouses in the winter and white flannel blouses in the summer, with long navy-blue skirts, linked to the blouses by elastic belts which continually slipped up or down, leaving exposed an unsightly hiatus of blouse-tape or safety-pinned shirt band.
The 'ZITA' Dip Front Adjuster features in _The Drapers' Record_ in May 1902 and is said to make the waist look smaller while giving 'that charming dip-front or long waist effect now so essential.' In 1909 the problem still existed and The Frank Patent Skirt Band was said to be the most comfortable way of achieving a 'straight-fronted effect'.
ACCESSORIES
In _Victorian Costume_ Anne Buck paid close attention to accessories as well as garments because she claimed that they were characteristic of the period. They were certainly an important element of the Edwardian woman's wardrobe; an ensemble would not have been complete without gloves – long for evenings and short for daywear. In 1905 Swan & Edgar's Ladies' Glove and Hosiery Department advertised day gloves in mocha suede or washing doeskin, while elbow-length evening gloves could be bought in kid or suede. A range of hose in black, tan or white was also available, held in place with elasticated garters trimmed with satin bows. In 1909 _The Lady's World_ advised that fashionable boots and shoes 'must always have the "flat pitch" toe, and are mainly black, but occasionally carried out with patent vamps and coloured doeskin uppers to match the dress in question.' To keep the evening chills at bay real fur stoles and muffs were worn, feather boas and neck trims made from marabou and ostrich were also popular. Advertisements in women's journals and trade catalogues show jewellery that was subtle and delicate: brooches worn at the front neck of a blouse, long strings of beads and pearls and contrasting, dangling earrings. There were also novelties in neckwear to pin to the front of a blouse or bodice. In April 1909 _The Drapers' Record_ ran an advertisement for John Howell & Co., Ltd, who produced a range of narrow satin jabots with dangling bobbles at the ends or tassels; muslin, lace and glacé cravats and simple poplin ties.
Postcard of a fashionably dressed woman wearing a range of accessories and carrying a macramé bag, 1912. (Jayne Shrimpton)
Hats were an important element of an Edwardian woman's daily dress and consequently there were hats for all occasions and hats priced to suit all incomes. Edwardian rules of etiquette dictated that the head must be covered outdoors and so for outdoor wear hats were a necessity for women of all classes. The more elaborate hats required anchoring in place by the use of a pair of long steel hat pins. The pins were pushed through the hair and into 'rats' – the pad used to create extra volume in a woman's hair. In Vita Sackville-West's novel, the following description is given of 'rats':
They were unappetising objects, like last year's birds-nests, hot and stuffy to the head...they provided the foundation on which the coiffure was to be swathed and piled, and into which the innumerable hairpins were to be stuck. It was always a great source of preoccupation with the ladies that no bit of the pad should show through the natural hair.
The leisured lady of fashion emphasized her status with large hats as witnessed in _Isobel's Dressmaking at Home,_ April 1901:
Hats and toques are all large, one may almost say immense. Fancy straws of the willow order compose many of the newest crowns, and tucked glacé silk is also largely employed, and the toque partakes of the nature of a turban with a deep brim and large crown.
Hat decorations continued to be lavish throughout the period. Inspiration was taken from nature and sometimes nature was directly added to hats, as in the case of birds. In April 1909 _The Drapers' Record_ noted that the latest hats from Paris were trimmed with large, green, outstretched birds' wings or even a half side of a bird in the most exquisite colours. _The Drapers' Record_ also noted, in 1909, the fashion in Paris for black hats, some huge, trimmed with bunches of roses that covered about a quarter of the hat. Flowers were popular hat decorations and at Shirley Leaf & Petal (the working artificial flower museum in the Sussex town of Hastings), catalogues and sample boards from 1910 show that small bunches of violets and large single roses were most in demand. Although millinery was a specialized occupation, women had the option to make their own hats. In the dressmaking journal _Weldon's Home Dressmaker_ Mrs Pritchard gives instructions on making a velvet toque – a turban-style hat. She writes:
Toques can be constructed out of almost any material, from tulle and lace for summer and evening wear to velvet, cloth and fur, which, when artistically draped, forms a most suitable and cosy headgear for winter.
Large felt hat with oversized plumage sketched by Ida Pritchard. (Worthing Museum and Art Gallery)
UNDERWEAR
The cleaning of a corset was a challenging process and so another layer that could be easily laundered was worn next to the skin, and this was the vest or chemise. The chemise was a long, sleeveless slip, which performed the function of a vest. Those observed in the museum collections appear to be made from cotton or linen. The better quality chemise might be embroidered around the neck with feather stich and have narrow, coloured ribbons threaded through broderie anglaise or lace. Clementina Black, in her survey of factory practices in _Married Women's Work,_ observed cheaper versions being made from flannelette or unbleached calico. Unbleached calico was described by one maker as 'the cheapest work' and she claimed it 'took more time to make up garments in it because the machine resisted the material. The chemise was worn tucked in to a pair of roomy drawers. Depending on the age and status of the wearer, Edwardian women had a choice of styles. Split drawers appeared in the early nineteenth century and although shortened in length, they were essentially just two separate legs reaching to the knee or sometimes beyond with fullness gathered onto a waistband that fastened at the back with ties. To ensure items of underclothing were always returned to their rightful owner after laundering, names and initials were sometimes elegantly embroidered on each piece; alternatively initials were crudely stitched in red thread.
Split drawers continued to be worn well into the Edwardian period. It is thought that younger women preferred closed drawers or knickers. Clementina Black observed women in London making drawers from flannelette and cotton. 'Knickers' was the term used by the women although Clementina Black stated that the garments were really drawers. In 1911, London department store D.H. Evans & Co were advertising 'French Peasant-made Longcloth Knickers, embroidered by hand' for sale at one shilling, eleven and a half pence. A further option was a pair of combinations, which were an all-in-one garment tapered to fit the upper half of the body. The advantage of combinations was the elimination of bulk around the waist.
With the underwear in place, the next layer would be a corset – this was something worn by all women, whatever their status. To go without a corset was very daring in an age defined by strict adherence to codes of etiquette. Lighter clothing often required an additional layer in the form of a camisole or corset cover. The camisole finished at the waist and was made from lightweight fabrics such as muslin or cotton batiste. They fastened down the front and were trimmed with lace and ribbons.
Worthing Museum has a selection of editions from the _Every-Woman's Encyclopaedia_ and although not dated it can be noted by the style of garments featured that they date from approximately 1910. Lace blouses were transparent and it would not have been proper for the wearer to exhibit her corset worn underneath, so a camisole, or corset cover was worn as an additional layer over the corset and under the lace or net blouse. The _Every-Woman's Encyclopaedia_ suggests a camisole with elbow-length sleeves as being particularly useful for wear under lace or transparent blouses. According to historian Alison Carter in her book _Underwear: The Fashion History,_ the term 'brassiere' first appeared in American _Vogue_ in 1907. As corsets began to drop downwards towards the waist the brassiere was introduced as a cover for the bare bosom rather than as a support to the bust. At this stage it did not have shaping and was a short loose tube with straps. Many were homemade and Worthing Museum has a pretty pink silk version with an embroidered motif and a large satin version made by an amateur dressmaker. C. Willett and Phillis Cunnington wrote in their book _The History of Underclothes,_ 'The Edwardian petticoat was always flimsy; not more than two were worn, the top one, particularly when coloured, often being referred to as an underskirt'. In 1905 Swan & Edgar advertised the 'FanFan', a smart silk underskirt with bone casings, accordion pleating and a top flounce finished with frills. By 1908 with the introduction of the narrower skirt, full petticoats were reduced to a less bulky form. A sketch by the fashion illustrator Ida Pritchard that features in Chapter 5 is an example of the slim-line petticoat.
OUTERWEAR
Blouses
Blouses were made from wool, cotton, linen, lace, and both natural and artificial silks. Edwardian blouses signified feminine respectability and satisfied strict codes of etiquette in relation to dress. They were highly adaptable and were suitable for daywear and sportswear whilst still retaining the elegance of high fashion, with stylistic features often originally dictated by Parisian designers. Blouses were priced to suit all incomes and were widely available to purchase at department stores, or could be made by home dressmakers. The first blouses began to appear in and around 1890. At first they were fitted and boned but by 1900 lighter fabrics were used and the boning had disappeared. Blouse styles evolved through the Edwardian period and by 1906 the predominant style was a pouched front, with three-quarter-length full sleeves and a high collar.
Dress historian Norah Waugh states that blouse collars started to be boned from about 1900 when lighter weight net and lace was used in a single layer and not backed with stiff muslin, which had been the case since the 1890s. By 1905 boned collars were usual in blouses and gowns with lace collars – in _Lucile Ltd: London, Paris, New York and Chicago, 1890s–1930s_ Valerie Mendes and Amy de la Haye wrote, 'Almost all the 1905 day wear is worn over blouses with tall, stand-up or stock collars that compelled wearers to hold their head up high.' Collar supports were necessary in order to keep lightweight lace collars upright and taut. Celluloid, which has the appearance of yellow plastic, was the most commonly observed support in blouse collars during this research. An example observed in a blouse in Worthing Museum has celluloid supports measuring 31/2 inches long with holes at either end to enable attachment by hand stitching. They were uncovered and would have been extremely uncomfortable to the wearer. An alternative, and often longer, collar support was a flattened wire spring. Occasionally they were wrapped with thread but this offered little protection and would still leave the wire support pressed against the neck ready to pinch folds of skin with any movement of the head. Metal rods were also used and were sometimes successfully covered in fabric. In some cases the fabric was not thick enough to prevent the sharp tips of the rods from tearing through, ready to make contact with the neck causing physical pain, as expressed by Cynthia Asquith. In her memoir, _Remember and Be Glad,_ she described the misery of wearing boned collars in the early 1900s: 'I couldn't endure the high choking collars with boned supports that dug red dints in my neck, so I wore low square-necked blouses long before these became the fashion – a nonconformity for which I was severely criticised.'
An example of metal collar supports used in a guipure lace blouse to keep the collar upright; one is missing and the other rusted support can be seen poking below the ribbon casing. (Worthing Museum and Art Gallery)
Tailored costumes and skirts
The tailored costume, otherwise known as the tailored suit, was a useful and practical outfit for the Edwardian woman. Ladies' tailors, court dressmakers and furriers John Redfern and Sons, based in Cowes on the Isle of Wight, were suppliers of quality, upmarket outerwear and women of means could purchase sporting dress, morning gowns and daywear. At the cheaper end of the market the Manchester-based firm John Noble were advertising 'half-guinea costumes' in the second edition of penny weekly _Home Chat_ published in 1895. The fitted bodice and matching skirt were 'the most remarkable value ever produced' at a price of 10s 6d. The skirt was also available to buy separately at 5s 6d. Although only available to buy in one choice of fabric, Cheviot Serge, the customer had a choice of colours: black, navy, brown, myrtle, bronze-green, tan, grey or drab. The advertising feature makes reference to the story of the only female survivor of a shipwreck to attest to the durability of its fabric. The young woman, Miss Anna Boecker, survived the sinking of the _Elbe,_ which went down very suddenly one morning in the North Sea. She was allegedly wearing a John Noble costume made from Cheviot Serge and despite being immersed in the water for five hours the costume had not shrunk. A comparison with newspaper reports at the time shows a different account: the captain of the ship that rescued the survivors spoke of seeing Miss Boecker lying on the floor of the lifeboat wearing only a long coat, having been pulled from the sea into the lifeboat after spending ten minutes in the freezing water.
Postcard (1910–1913) showing a woman in a two-piece costume wearing a large hat accessorized with a large flower. (Jayne Shrimpton)
Day dresses
Hand-painted fashion plate from Barrance and Ford showing a grey, black, white and brown day dress with a geometric trim applied around the hem of the skirt. (Peter Hinkins)
The day dress reproduced in Chapter 8 is from the Farebrother collection at Royal Pavilion & Museums, Brighton & Hove. Similar striped cotton summer dresses feature in many Edwardian fashion editorials. A further example can be seen in the design from the Brighton fashion house Barrance and Ford and in an illustration by Ida Pritchard. A pretty selection of summer day dresses made from white cotton lawn with lace insertions can be seen at Royal Pavilion & Museums; they are similar to lightweight blouses with lace insertions and tucks. Another lightweight option was the tea-gown. Dress historian Norah Waugh in her book _The Cut of Women's Clothes_ writes that in the period 1890–1908, 'Tea-gowns were loose and full, sometimes with "Watteau" back pleats or the Empire line, and with long, hanging sleeves.' Tea-gowns were worn in the afternoon when a woman wanted to feel comfortable at home and they were worn without a corset. The 'Empire' tea-gown was popular with _Isobel's Dressmaking at Home_ in 1901 – just the thing to step into at home to escape the cold March winds. The dress was constructed over a princess lining with a voile overdress with fullness arranged in a Watteau pleat at the back. An elbow-length sleeve with frilled edge would enable the gown to be 'perfectly finished'. The fullness of the overdress can be imagined when noting six yards of voile or eleven yards of soft silk was needed to make the tea-gown. 'Heliotrope' was the suggested colour, with black and white embroideries. Worthing Museum has a pink pleated wool tea-gown with cream lace inserts; it is a loose-fitting, front-fastening garment with full sleeves, resembling a loose negligée. In November 1907 Mrs Pritchard's weekly column in _The Ladies' Field_ advised readers of the importance of matching their tea-gown to their surroundings: the 'dark oak background and ample space' found in a large country house provided the perfect setting for a tea-gown of cerise, flame and orange. Mrs Pritchard appreciated the comfort provided by the tea-gown and accepted the lack of boning but preferred a tea-gown that defined the waistline. The tea-gown began to go out of fashion during the late Edwardian period; in 1909 _Lady's World_ wrote:
No one now wears the loose 'Empire' tea-gown, which every woman counted amongst her favourite, most comfortable garments not so long ago. The tea-gown in question was comfortable – there is no denying it – and it was often of beautiful material, daintily decorated. Its great drawback was its sloppy looseness, which utterly disguised the lines of the figure and which only very tall figures could carry well.
Evening gowns
Evening gowns were an ideal opportunity for displaying the Edwardian woman's love of embellishment. _The Ladies' Field_ reported the craze for a more fitted and structured Empire-line gown amongst wealthy socialites in January 1906. Gowns cut in this style had a high waistline that was slanted to be lower at the front and higher at the back. A jet-beaded bolero finishing at the same height was apparently a fashionable accessory to complement the Empireline gown in Monte Carlo.
Capes
Capes, cloaks and mantles were a useful way to wear an outer layer without disturbing and squashing the frills and flounces of the layer below. For Empire-line gowns a circular cape with an upturned 'half-moon' collar and low hood was suggested by the _Ladies' Supplement of the London Journal,_ because the hood could be pulled up over the head when walking to the carriage. An opera cloak was a longer and equally voluminous alternative with long, loose sleeves. Opera cloaks finished below the knee and skimmed over the gown underneath, often flaring out at the hem. In November 1906 _The Ladies' Field_ suggested pastel tones were desirable for an evening coat, made from wool satin with a silk lining and an interlining of domette for extra warmth.
Sporting clothes
In March 1889 _The Ladies' Field_ wrote of the revival of women's sports in the late nineteenth century. They attributed this to better diets and outdoor living. _The Ladies' Field_ contained a sports section – 'sports and pastimes' – a serious item of journalism that reported the statistics and achievements of women who excelled in their sport. Sporting dress was not mentioned in this section; readers had to consult the fashion pages for ideas. As sporting activities developed women wore and adapted their own clothing for playing outdoor sport but by the Edwardian period specific sports clothing was available to buy from department stores or could be made by the home dressmaker. Sports clothing design was based around the fashions of the day, although there were some concessions to comfort and ease of movement in the design of corsets.
Lawn tennis was popular with those who could afford the fees at membership-only tennis clubs. White was a popular colour for tennis dress because it disguised perspiration marks better than coloured fabrics. As well as the tennis dress, accessories were adapted and designed specifically for the game and the tennis racquet motif appeared on aprons with pockets used for carrying balls and on belts. The skirt and shirt combination was just the thing for the female golfer, especially when accessorized with a masculine-inspired tie and a simple beret.
WEARING THE CLOTHES AND ACCESSORIES IN THE BOOK
The garments and accessories in the book form a capsule collection – although this was not a term in use in the Edwardian period – with pieces that can be combined to create new outfits and adapted for a range of situations and characters. The outfits might be suitable for an Edwardian governess, a campaigning suffragette, a passenger on an ocean liner or a woman overseeing a busy household in large country house. The underwear – the drawers, chemise and petticoat – can be worn with all clothes with the exception of the evening gown. Patterns are cut for a modern shape but to achieve an authentic look a boned corset should be worn with all outfits, with the exception of the evening dress when a lighter corset can be worn. The white lace blouse can be worn with the plain skirt from the walking dress with the addition of a belt, or with the striped skirt from the day dress. The detachable black tie from the day dress can be worn at the front of the blouse collar. The velvet toque, a feather boa and a knitted or macramé Dorothy bag could be added to this ensemble to complete the look. The cape can be worn over the day dress or a combination of the blouse and a skirt. The day dress can also be accessorized with the lace parasol and the velvet toque; a pair of white, short, cotton lace gloves could also be worn. The beaded evening bag complements the evening gown. The evening gown can also be accessorized with long black gloves, black silk hose and a delicate headdress made from feathers and beads.
Suppliers
**Vena Cava** (suppliers of costumemaking equipment and parasol frames)
PO Box 3597, Poole, Dorset BH14 9ZL
Website: www.venacavadesign.co.uk
Email: info@venacavadesign.co.uk
**Ditto Fabrics**
21 Kensington Gardens, Brighton
BN1 4AL
Website: <https://dittofabrics.co.uk>
Phone: 01273 603771
**Fabric Land** (fabrics)
76 Western Road, Brighton, East Sussex
BN1 2HA
Website: www.fabricland.co.uk
Phone: 01425 461444
Email: maxine.fabricland@gmail.com
**Depotex** (fabrics)
16 Fisher Street, Lewes BN7 2DG
Phone: 01273 487956
**Merchant and Mills** (fabrics and sewing equipment)
14A Tower Street, Rye, East Sussex
TN31 7AT
Website: <http://merchantandmills.com>
Phone: 01797 227789
**Simply Sequins** (online sequin shop)
82 Durrants Road, Rowlands Castle,
Hampshire PO9 6BG
Website: www.simplysequins.co.uk
Phone: 023 9247 6125
Email: sales@simplysequins.co.uk
**Wayward** (new and vintage fabrics and trimmings)
68 Norman Road, St Leonards on Sea,
East Sussex TN38 0EJ
Website: <http://wayward.co>
Phone: 07815 013337
Email: info@wayward.co
**Jaycotts** (sewing equipment)
Unit D2, Chester Trade Park, Bumpers
Lane, Chester, Cheshire CH1 4LT
Website: www.jaycotts.co.uk
Phone: 01244 394099
Email: alex@jaycotts.co.uk
**William Gee** (pattern cutting paper and sewing equipment)
520–522 Kingsland Road, London
E8 4AH
Website: www.williamgee.co.uk
Phone: 0207 254 2451
Email: info@williamgee.co.uk
**The Bead Shop Brighton**
21 Sydney Street, Brighton BN1 4EN
Website: <http://beadsunlimited.co.uk>
Phone: 01273 675077
Email: shop@beadsunlimited.co.uk
**Anne Tomlin** (flower artist, makes flowers to order and runs classes in flower making; also gives individual tuition)
Email: aetomlin@gmail.com
**Lomax and Skinner** (milliners who make bespoke hats, run classes and give individual tuition at their workshop)
66 The High Street, Lewes, East Sussex
BN7 1XG
Website: www.lomaxandskinner.co.uk
Phone: 07949 123693
Email: studio@lomaxandskinner.co.uk
**Morplan** (sewing equipment)
Unit 1, Temple Bank, Harlow, Essex
CM20 2DY
Website: www.morplan.com/shop/en/morplan
Phone: 0800 451122
Email: web.support@morplan.com
Bibliography
BOOKS
Arnold, Janet (1982). _Patterns of Fashion 2: Englishwomen's Dresses and Their Construction_ c. _1860–1940_. Macmillan.
Asquith, Cynthia (1952). _Remember and Be Glad._ Charles Scribner's Sons.
Black, Clementina (1915). _Married Women's Work._ London: G. Bell and Sons.
Brittain, Vera (1978). _Testament of Youth._ Virago.
Buck, Anne (1984). _Victorian Costume._ Ruth Bean Publishers.
Campbell, Lady Colin (1912). _Etiquette of Good Society._ Cassell.
Carter, Alison (1992). _Underwear: The Fashion History._ London: B.T. Batsford Ltd.
Earnshaw, Pat (1982). _A Dictionary of Lace._ Shire Publications.
Hopkins, J.C. (1990). _Edwardian Ladies' Tailoring: The Twentieth Century System of Ladies' Garment Cutting_ (1910). R.L. Shep.
Hunnisett, Jean (1988). _Period Costumes for Stage and Screen: Patterns for Women's Dress 1800–1909_. Unwin Paperbacks.
Mendes, Valerie D. and de la Haye, Amy (2009). _Lucile Ltd: London, Paris, New York and Chicago_ , _1890s–1930s._ V&A Publishing.
Meyer, Mrs Carl and Black, Clementina (1909). _Makers of our Clothes._ Duckworth & Co.
Sackville-West, Vita (1983). _The Edwardians._ Virago Modern Classics.
Smith, Amy K. (1910). _Cutting Out for Student Teachers._ Sir Isaac Pitman & Sons Ltd.
Taylor, Lou (2002). _The Study of Dress History._ Manchester University Press.
Walker, Agnes (1907). _How to Make up Garments._ Blackie & Sons Ltd.
Waugh, Norah (1994). _The Cut of Women's Clothes: 1600–1930._ Faber and Faber.
Willett C. and Cunnington, Phillis (1992). _The History of Underclothes._ Dover Publications Inc.
Woman's Institute of Domestic Arts & Sciences, Scranton, PA:
_Embroidery and Decorative Stitches_ (1916)
_The Dressmaker and Tailor's Shop_ (1917)
_Underwear and Lingerie_ (1921)
_Essential Stitches and Seams_ (1922)
_Tailored Seams and Plackets_ (1923)
PERIODICALS
_The Ladies' Field_
_Home Chat_
_The Drapers' Record_
_The Lady's World_
_Isobel's Dressmaking at Home_
_Weldon's Home Dressmaker_
_Weldon's Illustrated Dressmaker_
_Weldon's Practical Needlework_
_The Queen_
_Woman's World_
MUSEUMS VISITED FOR RESEARCH
Royal Pavilion & Museums, Brighton & Hove
Worthing Museum and Art Gallery
Museum of London
The Clothworkers' Centre for the Study and Conservation of Textiles
Acknowledgements
This book would not have been possible without the kindness and support of Martin Pel, Curator of Fashion and Textiles at Royal Pavilion & Museums, Brighton & Hove, and Gerry Connolly, Curator of Historic Collections at Worthing Museum and Art Gallery. Both curators gave generous access to the wonderful collections at their respective museums and shared their expertise. I am also grateful to Professor Lou Taylor who first inspired my interest in Edwardian fashion by guiding me towards the production and consumption of Edwardian blouses, which became the subject of my MA dissertation. I am grateful to the following museums and archives for allowing research at their collections: Museum of London, University of Brighton Design Archives, London College of Fashion Archive and The Clothworkers' Centre for the Study and Conservation of Textiles.
I am indebted to costume designer Frances Tempest and theatrical tailor and cutter Tony Rutherford for sharing professional tips, and Stephanie Richards for reading my introduction and offering comments. Many thanks to Jayne Shrimpton and to Katharine Williams for supplying Edwardian photographs and giving permission to publish, and to Peter Hinkins for giving permission to reproduce fashion plates from the Barrance and Ford, Brighton Ltd, catalogue. I am also grateful to EMap for allowing the publication of images from _The Drapers' Record_ Archive, and to Worthing Museum and Art Gallery for permission to publish images from the collection of fashion illustrator Ida Pritchard and from their collection of dressmaking manuals and journals. Further thanks are due to Marion May who kindly showed me her Edwardian dress collection and gave permission for the reproduction of images.
I am very grateful to Karen Cunningham for skilfully cutting the finished patterns and to skilled costume maker Kate Stallion for helping with costume making. Many thanks to costumier Rachel Woolcott, who made such a good job of beading the evening bag and helped to make the split drawers. Jo Lance, design history graduate and milliner, did a great job of making the velvet toque, and the beautiful flower was researched and made by flower artist Anne Tomlin. I would also like to thank Simon Barclay at Depotex fabric shop in Lewes for advising on fabrics.
I am grateful to Benjamin Rowland for taking photographs for chapter openers and for editing my photographs. Thank you also to photographer Andrew Perris who photographed the finished underwear and walking dress. Many thanks to illustrator Joe McCrae who sketched the original museum garments and to Zachary Rowland who sketched the measurements guide.
Finally, I would like to thank Chris Rowland who adapted his CAD drawing skills to reduce the patterns, and provided endless cups of tea and support.
Index
accessories
making 104–117
wearing 120–122
Asquith, Cynthia, Edwardian memoirs , 57–58, 122–123
alterations and repairs
beading
beading techniques , 91–95, 105–108
evening bag 105–107
evening gown 92–95
beeswax
bobbles
bodkin
boning
brush braid
buttons and buttonholes, two-part walking dress bodice
chemise 36–39
collars
blouse, lace collar 54–55
cape 99–103
collar supports , , 122–123
day dress, lace collar 77–78,
two-part walking dress 66–67
cutting and layout
blouse 50–51
cape 98–99
chemise
day dress 69–73
evening bag
evening gown 85–89
flower motif, evening gown
neck tie, day dress
parasol
petticoat 43–44
split drawers 34–35
two-part walking dress bodice
two-part walking dress, skirt 58–59
velvet toque 109–110
darts
cape, open darts
petticoat, closed darts
department stores offering made-to-measure, history 10–12
drafting paper
dress stand
Edwardian dressmaking history 12–13,
Edwardian silhouette
Empire tea-gown, history 123–124
epaulets and strappings, day dress 76–77
evening bag, making 105–108
fabrics
Edwardian colours and prints 25–26
glossary of fabrics 26–29
sourcing fabrics
faggoting
Farebrother, Katherine, Edwardian clothing
fittings
flounce
flounce for cape
flounce for petticoat ,
foundation bodice, evening gown 86–87
Fray Check
frill, petticoat ,
gathering, chemise
Gibson Girl
glossary of fabrics 26–29
grain lines
hats, history 120–121
hat making project, see toque
hem facings
day dress, bias cut hem facing
two-part walking dress, shaped hem facing
hooks and bars, day dress
insertion lace
interfacing and boning
measurements guide ,
millinery wire hat frame, making 110–111
museum research
neck tie
needles for hand sewing
Noble, John, half-guinea costumes
parasol, making 113–117
parasols, Worthing Museum 113–114
patterns
adding seam allowance
Edwardian patterns, history 19–20
enlarging ,
grain lines
making a toile to test fit
sizes
taking patterns from existing garments
peplum, day dress
pin-tucks
blouse 52–53
petticoat
plackets ,
pockets
cape patch pocket
skirt patch pocket, two-part walking dress 62–63
skirt placket hole pocket, two-part walking dress 59–60
skirt watch pocket, two-part walking dress
pressing tools and techniques
Pritchard, Mrs Eric, Edwardian fashion journalist ,
Pritchard, Ida, Edwardian fashion illustrator , , ,
quilted cape lining
rolling and whipping
rouleau loop
ruffles, cape 100–101
Rutherford, Tony, tips from a theatrical cutter and tailor
scarf, evening gown
scissors 16–17
seam allowance, adding to patterns ,
seams
Edwardian seams
faggoting
flat felled ,
French ,
open
sleeves
blouse sleeve with inserts
draped, evening gown ,
setting in 65–66,
two-piece, two-part walking dress 64–65
sewing machines
side panels, evening gown
silhouette ,
skirt band, for holding a blouse inside a skirt 119–120
Smith, Amy K., _Cutting Out for Student Teachers_ (1910)
split drawers 33–36
sporting clothes, history
tacking techniques 20–21
tape measure
Tempest, Frances, tips from a costume designer
thimble
toque 109–113
tracing wheel
trimmings
tucks
blouse 52–53
chemise 37–38
day dress 73–74
two-part walking dress 66–67
underskirt, evening gown 90–91
underwear
how underwear was worn , 121–122
making drawers 33–36
making the chemise 36–39
waistbands
day dress
Petersham
split drawers 35–36
two-part walking dress
waist stay ,
Walker, Agnes, _How to Make Up Garments_ (1907)
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 1,877
|
Q: Plotting non-injective labels on vertices in a graph in SageMath I have a graph with vertices labeled by some pairs (a,b).
Can I plot it in such a way that I only see the first component "a" printed over each vertex?
I cannot just relabel, since my map (a,b)->a is not injective.
For a small example, take
G = Graph()
G.add_edge((1,1),(1,2))
The usual G.plot() gives (1,1)---(1,2).
Instead, how to produce only 1---1 ?
A: Plot a Sage graph with non-injective vertex labals
We describe a slightly tedious workaround, and then
a way to recover our comfort.
New class for vertices
One solution consists in writing a new class for vertices
that inherits from tuple and has a custom __str__ method
that returns a string for only the first entry in the tuple.
class MyVertex(tuple):
r"""
Class for vertices for special plotting of graphs.
Use with tuples, and only the first entry in the tuple
will be used as a vertex label when plotting the graph.
"""
def __init__(self, v):
self.vertex = v
def __str__(self):
return str(self.vertex[0])
Using this to define the vertices of the graph,
we obtain the desired behaviour.
Define a graph and add an edge from (1, 1) to (1, 2):
sage: G = Graph()
sage: G.add_edge(MyVertex((1, 1)), MyVertex((1, 2)))
When plotting the graph, both vertices have the label 1.
sage: G.plot()
Launched png viewer for Graphics object consisting of 4 graphics primitives
When listing the vertices, they still appear fully:
sage: G.vertices()
[(1, 1), (1, 2)]
Using usual graphs
To avoid having to use MyVertex explicitly, we write a graph-plotting
function that creates an intermediate "MyVertex"-style copy of a usual graph
for the sake of plotting.
def plot_graph(G):
r"""
Return a plot of this graph with special vertex labels.
The graph vertices are assumed to be tuples. The plot
uses the first component of each tuple as a vertex label.
"""
E = [(MyVertex(a), MyVertex(b)) for (a, b) in G.edges(labels=False)]
return Graph(E).plot()
Now compare:
sage: G = graphs.Grid2dGraph(3, 4)
sage: G.plot()
Launched png viewer for Graphics object consisting of 30 graphics primitives
sage: plot_graph(G)
Launched png viewer for Graphics object consisting of 30 graphics primitives
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 5,226
|
\section{Introduction}
\label{sec:intro}
\begin{figure}[!t]
\begin{center}
{\includegraphics[width=1.0\linewidth]{figs/Fig1.pdf}}
\end{center}
\vspace{-0.5cm}
\caption{
(a) Pure 3D medical SSL learns representations with only 3D CT scans;
(b) our proposed UniMiSS brings a wealth of 2D X-rays to offset the lack of 3D data, thus enables the large-scale SSL for better pre-training performance. Besides, the pre-trained model is generic to various downstream (DS) applications, without the restriction on the dimensionality barrier.
}
\label{fig:fig1}
\vspace{-0.3cm}
\end{figure}
Medical image analysis, a key process in computer-aided diagnosis, is well known by its lack of labels for training, especially for the 3D task.
Recent research work suggests that the self-supervised learning (SSL) is promising to ease the annotation cost by making the best of unlabeled data~\cite{GLcontras,SSL_MedIA_resto,MoCoXray,3Dssl_Taleb,PGL,PCRL,3Dssl_zhou,SSL_MedIA_cube}.
Although setting label free, SSL still heavily relies on the large-scale unlabeled data to explore the feature representations.
Unfortunately, publicly available 3D medical data is relatively limited due to the high imaging cost and data privacy.
Most of 3D medical datasets just contain a few thousands of cases.
For example, Zhou~\etal~\cite{PCRL} utilized the LUNA dataset~\cite{LUNA16}, containing about 1000 CT cases, for self-supervised pre-training.
Such a small data scale may limit the potential of SSL in 3D medical image analysis.
In comparison to 3D data, it is easy to collect hundreds of thousands of 2D medical images such as X-rays due to its fast imaging speed, low radiation and low cost.
Accordingly, we advocate to bring a wealth of 2D medical images to the 3D SSL process, aiming at learning strong representations with large-scale images, as shown in Fig.~\ref{fig:fig1}. Comparing to the pure 3D medical SSL, this practice benefits the medical SSL in terms of three significant merits.
First, 2D data serves as a compensation for the lack of 3D data, enabling the large-scale SSL pre-training.
Second, there is the anatomy correlation between 2D and 3D images, like chest X-ray and CT. Such an intrinsic relevance may contribute for strong associated representations.
Third, the pre-trained model is generic enough to be applied to both 3D and 2D downstream tasks.
To achieve the universal SSL purpose, on the technical side, we need to build a versatile model that is able to process both 2D and 3D images.
The common practice in medical image analysis is to design 2D convolutional neural networks (CNNs) for 2D images~\cite{Xie_TMI_skin,Zhang_TMI_ano,PCRL} and 3D CNNs for 3D images~\cite{PGL,CoTr,dodnet,PCRL,3Dssl_zhou}, respectively.
Restricted to the dimensionality barrier, it is almost impossible to design a dimension-free CNN network for this purpose.
Recent months have witnessed the success of Transformer in computer vision~\cite{ViT}. A vision Transformer usually takes a sequence of image patches, represented by the learned linear embedding, as the input to model the long-term dependencies among the sequence elements.
Owing to the sequence modeling, Transformer can accept the data of any dimensions, including but not limited to 2D images and 3D volumetric data. Therefore, Transformer offers the possibility of breaking the dimensionality barrier and constructing a universal SSL model.
In this paper, we propose a \textbf{Uni}versal \textbf{M}ed\textbf{i}cal \textbf{S}elf-\textbf{S}upervised representation learning framework (UniMiSS) that learns general representations from 2D and 3D unlabeled medical images.
To achieve this, we design a dimension-free pyramid U-like \textbf{M}ed\textbf{i}cal \textbf{T}ransformer (MiT), which is mainly composed of switchable patch embedding (SPE) module and Transformers.
The SPE module converts the input images to a sequence by using 2D or 3D patch embedding, depending on the input dimension.
The Transformer layer processes the embedded tokens in a sequence-to-sequence manner, regardless of their original dimension.
We perform the self-supervised learning by the self-distillation of student and teacher networks, both of which take the MiT as the backbone.
The student network learns to predict the output distribution obtained with the momentum teacher network, following the view consistency.
Moreover, the 3D volumetric image should be identical with their slices due to the same imaging content.
The volume-slice consistency is adopted as a cross-dimension regularization to boost the representations.
We conduct the SSL experiments based on 5,022 3D CT volumes, which are augmented by 108,948 2D X-ray images.
Benefit from the huge augmented 2D data, the proposed UniMiSS achieves the obvious performance improvement on the downstream 3D classification/segmentation tasks.
Besides, the UniMiSS pre-trained model can be freely applied to 2D downstream tasks, which beats strong competitors like ImageNet pre-training on the downstream 2D medical tasks.
To summarise, our contributions are three-fold:
(1) we are the first to augment 3D medical images with the easily accessible unpaired 2D ones for the SSL purpose, aiming at addressing the limitation of 3D data amounts during the SSL process;
(2) the proposed MiT breaks the dimensionality barrier and enables the joint SSL training with both 2D and 3D images;
and (3) our UniMiSS pre-training achieves the advanced performance on six downstream tasks, covering the 3D/2D medical image classification/segmentation.
\section{Related Work}
\label{sec:relatedwork}
\subsection{Self-supervised Learning}
SSL has been extensively studied in the literature. According to the pretext tasks, these studies can be broadly categorized into the discriminative methods~\cite{Clustering,SimCLR,BYOL,MoCo,DIM,Transformation,PIPL,Jigsaw,CPC,CMC} and generative methods~\cite{Colorization,Super_resolution,Inpainting,GAN,Decoupling}. The contrastive learning~\cite{SimCLR,BYOL,MoCo,DIM,PIPL,CPC,CMC} has drawn significant research attention and achieved advanced performance on many vision tasks.
Most of the previous work were built on the CNN-based network.
More recently, Transformer has become an increasingly popular alternative architecture in computer vision.
There has been a trend towards combining the merits of Transformer and SSL, advancing the self-supervised vision Transformers.
The seminal work is iGPT~\cite{iGPT}, which follows the masked auto-regressive language modelling to pre-train the self-supervised vision Transformer.
Besides, some attempts have also been made to pre-train vision Transformers using the contrastive learning~\cite{DINO} or Siamese distillation~\cite{MoCov3}, which outperform the CNN-based SSL approaches, setting a new record on ImageNet.
The success of SSL in computer vision also benefits to the medical community~\cite{GLcontras,SSL_MedIA_resto,MoCoXray,3Dssl_Taleb,PGL,PCRL,3Dssl_zhou,SSL_MedIA_cube}. Typical attempts include pre-training a CNN by restoring the content of raw images~\cite{SSL_MedIA_resto,3Dssl_Taleb,PCRL,3Dssl_zhou,SSL_MedIA_cube} and tailoring contrastive SSL to medical images~\cite{GLcontras,MoCoXray,3Dssl_Taleb,PGL}.
These efforts constitute an important and timely step forward towards better SSL approaches to medical image analysis. However, they suffer two limitations.
First, the CNN architecture enables the pre-training on either 2D or 3D medical images, failing to process both of them simultaneously. The resulting representations would be trapped especially for the limited 3D data.
Consequently, the pre-trained CNN can only be transferred to the dimension-specific downstream task.
Second, the above SSL approaches capture the spatial context of 3D medical images from either slices~\cite{GLcontras} or volume~\cite{3Dssl_Taleb,3Dssl_zhou,SSL_MedIA_cube}. Few of them consider the inherent consistency relation between volume and its slices.
\subsection{Cross-Domain Training for Medical Imaging}
In the medical context, the cross-domain training usually jointly utilizes two or more datasets acquired at different sites~\cite{CrossD_MICCAI18,CrossD_Liu} or using different imaging modalities~\cite{CrossD_Dou,CrossD_Kli,CrossD_CVPR18} to train a single model that could perform well on diverse datasets.
Karani~\emph{et~al.}~\cite{CrossD_MICCAI18} and Liu~\emph{et~al.}~\cite{CrossD_Liu} trained a single CNN with shared convolutional layers and specific batch normalization layers using the MRI data acquired at each site individually, aiming to tackle the statistical divergence explicitly.
Zhang~\emph{et~al.}~\cite{CrossD_CVPR18} simultaneously learned a volume-to-volume translation using the unpaired CT and MRI data and strong segmentors using synthetic data, which were translated from another modality.
Dou~\emph{et~al.}~\cite{CrossD_Dou} derived a variant of knowledge distillation (KD) to leverage the shared across-modality information between CT and MRI for accurate segmentation of anatomical structures.
Li~\emph{et~al.}~\cite{CrossD_Kli} also introduced KD to the cross-modality analysis of CT and MRI data, but they simultaneously exploited abundant unlabeled data.
These studies are dedicated to analyzing multi-modal/site but fixed dimension (3D) medical images, failing to address the dimensionality barrier in our scenario.
\section{Methods}
\label{sec:method}
\begin{figure*}[!t]
\begin{center}
{\includegraphics[width=1.0\linewidth]{figs/Fig2.pdf}}
\end{center}
\vspace{-0.5cm}
\caption{Illustration of the proposed UniMiSS framework. It has a dual path architecture, \ie, a student and a teacher. Taking both 2D X-rays and 3D CTs as input, UniMiSS is trained by the self-distillation strategy, \ie, maximizing the agreement of both paths. To break the dimensionality barrier between X-rays and CTs, the MiT network, composed of the switchable patch embedding (SPE) module and Transformers, processes the 3D/2D data in a sequence-to-sequence manner.
}
\label{fig:fig2}
\vspace{-0.1cm}
\end{figure*}
\subsection{Overview}
UniMiSS is a universal medical SSL framework that is superior to learn general image representations with large scale mixed 2D and 3D unlabeled medical images.
Figure~\ref{fig:fig2} illustrates the pipeline of UniMiSS.
Let us denote the mixed 2D and 3D data pool by $\{\mathbb{D}^{2D}, \mathbb{D}^{3D}\}$.
To enable UniMiSS to process both 2D and 3D medical images, we build the MiT as its backbone, which is mainly constituted by the dimension-adaptive SPE module and Transformer layers.
We perform the SSL process in the self-distillation manner, and utilize a standard cross-entropy loss to maximize the consistency between the student and teacher outputs. Besides, to get the utmost out of 3D volumetric information, we introduce the volume-slice consistency constraint, which encourages UniMiSS to model the consistency cross dimensions. It is intuitively conducive to learning strong feature representations from the volumetric images. We now delve into the details of this framework.
\subsection{MiT: A Dimension-free Architecture}
Although achieving great success in computer vision, vision Transformer~\cite{ViT} still remains challenging to process high resolution 3D images, due to the high computation cost and memory requirement.
Inspired by~\cite{PVT}, we design the MiT with a pyramid architecture to process both 2D and 3D images efficiently.
To break the dimensionality barrier, we propose a simple yet efficient SPE module to adaptively choose the 2D or 3D patch embedding according to the input type.
MiT has an encoder-decoder architecture that facilitates the various applications, including segmentation and classification.
We now describe each part of MiT, and more details can be found in Appendix.
\noindent
\textbf{SPE.}
As shown in Figure~\ref{fig:fig2}, the SPE module plays an important role to obtain the dimension-specific embedding, \ie, using 2D patch embedding operation for 2D inputs and using 3D patch embedding operation for 3D inputs
. Notice that the implementations of SPE in the encoder and decoder are different. The SPE in the encoder refers to a switchable 2D and 3D convolution block with the stride of 2, which reduces the feature resolution. In contrast, the SPE in the decoder is a switchable 2D and 3D transpose convolution block, which increases the feature resolution.
\noindent
\textbf{Encoder-Decoder.}
The MiT encoder follows a progressive shrinking pyramid Transformer, as done in ~\cite{PVT}. It consists explicitly of four stages, each of which is composed of a SPE module and several stacked Transformers.
In each stage, the SPE module down-samples the input features and generates the dimension-specific embedded sequence.
Notably, we append an extra learnable SSL token~\cite{DINO,MoCov3} to the patch embedded sequence.
The SSL token is similar to the [CLS] token in ViT, which is able to aggregate information from the whole patch embedding tokens via the self-attention.
The resultant sequences, combined with the learnable positional embedding, are inputted into the following Transformers for the long-term dependency modeling.
Each Transformer layer includes a self-attention module and a feed-forward network (FFN) with two hidden layers.
To enable MiT to process high-resolution images, we follow the spatial-reduction attention (SRA) layer~\cite{PVT}.
Given a query $\bm{q}$, a key $\bm{k}$, and a value $\bm{v}$ as the input, SRA first reduces the spatial resolution of $\bm{k}$ and $\bm{v}$, and then feeds $\bm{q}$, reduced $\bm{k}$, and reduced $\bm{v}$ to a multi-head self-attention (MSA) layer to produce refined features. This process can be formally expressed as follows
\begin{equation}
SRA(\bm{q},\bm{k},\bm{v}) = MSA(\bm{q}, F(\sigma(R(\bm{k}))), F(\sigma(R(\bm{v})))),
\end{equation}
where $\sigma(\cdot )$ represents a linear projection, \ie, strided 2D or 3D convolution operation, that reduces the feature map resolution, $R(\cdot )$ reshapes the input sequence to a feature map of the original spatial size, and $F(\cdot )$ flattens the input into a 1D sequence.
MiT has a symmetric decoder structure that consists of three stages. In each stage, the input feature map is first up-sampled by the SPE module, and then refined by the stacked Transformer layers.
Besides, we also add skip connections between the encoder and decoder to keep more low-level but high-resolution information.
\subsection{Objective of UniMiSS}
The proposed UniMiSS framework is based on the student-teacher paradigm.
Each path comprises a MiT network $\mathcal{F}_\theta(\cdot)$ and a projector $\mathcal{P}_\theta(\cdot)$.
$\mathcal{P}_\theta(\cdot)$ is a $n$-layer multi-layer perceptron (MLP) head, $\theta$ represents the parameter set of this path.
The SPE layers switch to perform the 2D patch embedding or 3D patch embedding during the feed-forward computing that is denoted as $\mathcal{F}_\theta(\cdot; 2D)$ and $\mathcal{F}_\theta(\cdot; 3D)$, respectively.
During the SSL process, we only extract the SSL token from the output of $\mathcal{F}_\theta(\cdot; 2D/3D)$ as the input of the projector.
Since the Transformer sets the dimension free, our UniMiSS is able to learn image representations from both 2D and 3D unlabeled medical images.
Both of paths share an identical architecture. However, they differ in the following two items. First, the teacher network is formulated as a momentum version of the student network, which updated by an exponential moving average strategy, defined as
\begin{equation}
\mu \leftarrow \lambda\mu+(1-\lambda)\theta,
\end{equation}
where $\lambda$ increases from 0.996 to 1 using a cosine schedule during training~\cite{DINO}.
Second, a stop-gradient operator is performed to the teacher network to avoid model collapse.
\noindent
\textbf{Objective for 2D domain data.}
Taking a mini-batch of 2D data $\bm{x}$ for example, we first create two augmented views $\bm{x}_1$ and $\bm{x}_2$ by using the data augmentation module $\mathcal{T}$, and then feed them into the student and teacher networks. The obtained SSL token is inputted into the projector to produce the output vector, denoted as $\bm{f}_1 = \mathcal{P}_\theta(\mathcal{F}_\theta(\bm{x}_1; 2D))$, $\bm{f}_2 = \mathcal{P}_\mu(\mathcal{F}_\mu(\bm{x}_2; 2D))$.
The objective of UniMiSS is to maximize the consistency between the output vectors obtained with student and teacher networks, formulated by
\begin{equation}
\mathcal{H}(\bm{f}_1, \bm{f}_2)=-\mathrm{softmax}(\frac{\bm{f}_2-\mathcal{C}}{\tau_t}) * \log(\mathrm{softmax}(\frac{\bm{f}_1}{\tau_s})),
\end{equation}
where $\mathcal{C}$ is the centering of teacher outputs, $\tau_t$ and $\tau_s$ are sharpening temperature parameters for student and teacher network.
The centering operation heartens the model to the uniform distribution while the sharpening has the opposite effect, \ie, encouraging one dimension to dominate. Both of them are jointly used together to avoid model collapse~\cite{DINO}.
Specifically, the temperature $\tau_t$ is set to a small value in the teacher path for the sharpening purpose.
The center $\mathcal{C}$ is first computed via averaging the teacher's outputs of the min-batch data and then updated with an exponential moving average strategy to aggregate the center across the whole batches, shown as follows
\begin{equation}
\mathcal{C} \leftarrow \omega * \mathcal{C}+(1-\omega )*\widehat{\bm{f}_2}
\end{equation}
where $\omega$ is a rate parameter, and $\widehat{\bm{f}_2}$ refers to the mean of teacher output in a mini-batch.
We define a symmetrized loss for 2D images as:
\begin{equation}
\mathcal{L}^{\mathrm{2D}}=\mathbb{E}_{\bm{x}\sim \mathbb{D}^{2D}}[\mathcal{H}(\bm{f}_1, \bm{f}_2) + \mathcal{H}(\bm{f}_2, \bm{f}_1)]
\end{equation}
\noindent
\textbf{Objective for 3D domain data.}
In medical domain, 3D volumes can be viewed as the stacking of 2D images along with the inter-slice dimension. The volume data has the inherent consistency to their slices, which inspires us to model the volume-slice consistency for SSL.
Given a 3D data $\bm{x}$ sampled from the 3D medical dataset, we denote its two augmented views as $\bm{x}_1$ and $\bm{x}_2$, each containing $m$ 2D slices.
We compute the global volumetric representations by the student and teacher networks in a 3D mode, \ie, $\bm{f}_1 = \mathcal{P}_\theta(\mathcal{F}_\theta(\bm{x}_1; 3D))$, and $\bm{f}_2 = \mathcal{P}_\mu(\mathcal{F}_\mu(\bm{x}_2; 3D))$.
Meanwhile, we stack $m$ slices of each augmented view in a batch, and use them as 2D inputs to calculate the slice-wise representations in a 2D mode, and then treat the average outputs of all slices as the holistic slice representations, \ie, $\bm{f}'_1=\frac{1}{m}\sum_{i=1}^{m}\mathcal{P}_\theta(\mathcal{F}_\theta([\bm{x}_1]^i; 2D))$, and $\bm{f}'_2=\frac{1}{m}\sum_{i=1}^{m}\mathcal{P}_\mu(\mathcal{F}_\mu([\bm{x}_2]^i;2D))$, where $[\bm{x}]^i$ represents the $i$-th slice extracted from the 3D data $\bm{x}$.
After that, we build the following objective function
\begin{equation}
\begin{split}
\mathcal{L}^{\mathrm{3D}}=\mathbb{E}_{\bm{x}\sim \mathbb{D}^{3D}}&[\mathcal{H}(\bm{f}_1, \bm{f}_2) + \mathcal{H}(\bm{f}_1, \bm{f}'_2) + \mathcal{H}(\bm{f}'_1, \bm{f}_2) + \mathcal{H}(\bm{f}'_1, \bm{f}'_2) \\
&+\mathcal{H}(\bm{f}_2, \bm{f}_1) + \mathcal{H}(\bm{f}_2, \bm{f}'_1) + \mathcal{H}(\bm{f}'_2, \bm{f}_1) + \mathcal{H}(\bm{f}'_2, \bm{f}'_1)]
\end{split}
\end{equation}
The above objective function encourages to learn the refined consistency with 3D medical data in terms of three aspects, \ie, volume to volume, slice to slice, and volume to slice.
We introduce an alternative training scheme to solve this multi-objective optimization problem.
As shown in Figure~\ref{fig:fig3}, we first sample 2D images to train the UniMiSS from step 0 to step $\upsilon$, and then take turn to sample 3D volumes in the next $\upsilon$ steps. The following training process will continue in a circular manner until the model converges.
The proposed iterative training scheme has two merits: (1) it bypasses the difficulty of using both 2D and 3D images in the same batch; and (2) it can reduce the instability caused by the distribution discrepancy between 2D and 3D data.
\begin{figure}[!t]
\begin{center}
{\includegraphics[width=0.7\linewidth]{figs/Fig3.pdf}}
\end{center}
\vspace{-0.5cm}
\caption{Illustration of 2D-and-3D alternation training.}
\label{fig:fig3}
\vspace{-0.1cm}
\end{figure}
\section{Experiments}
\label{sec:Experiments}
\subsection{Datasets}
\noindent
\textbf{Pre-training datasets.}
We collected 5,022 3D CT scans from five datasets
(\ie MOTS dataset~\cite{dodnet},
LIDC-IDRI dataset~\cite{LIDC},
Tianchi dataset~\cite{Tianchi},
RibFrac dataset~\cite{RibFrac},
TCIACT dataset~\cite{TCIACT}),
and collected 108,948 2D images from NIH ChestX-ray8 dataset~\cite{Chestxray8} to train UniMiSS in a self-supervised manner.
\noindent
\textbf{Downstream datasets.}
Table~\ref{tab:tab1} gives the details of six downstream tasks, which can be grouped into (1) 3D downstream: CT-based segmentation (BCV) and classification (RICORD), MRI-based segmentation (CHAOS); (2) 2D downstream: multi-organ segmentation (JSRT) and pneumonia classification (ChestXR), and skin lesion segmentation (ISIC).
Note that the CHAOS and ISIC datasets are different from the pre-training data in terms of modalities (\ie, 3D CT vs. MRI, 2D X-ray vs. dermoscopy). They are used to evaluate the unseen-modality transferability.
\begin{table}[t!]
\footnotesize
\caption{Six datasets for the downstream evaluation. Noticed that we used two test sets, \ie offline test set (off) and online test set (on), for the BCV dataset.}
\label{tab:tab1}
\vspace{-0.5cm}
\begin{center}
\begin{tabular}{c|c|c|c|c}
\hline
\multicolumn{5}{c}{Downstream evaluation datasets} \\ \hline
Name & Tasks & Modalities & \#Train & \#Test\\ \hline
BCV~\cite{BCV} & Multi-organ segmentation & \multirow{2}{*}{3D CT} & 24 & 6 (off)+20 (on)\\ \cline{1-2} \cline{4-5}
RICORD~\cite{RICORD} & COVID-19 screening & & 182 & 45\\ \hline
CHAOS~\cite{CHAOS} & Abdominal organ segmentation & 3D MRI & 48 & 12\\ \hline
JSRT~\cite{JSRT,SCR} & Multi-organ segmentation & \multirow{2}{*}{2D X-ray} & 124 & 123\\ \cline{1-2} \cline{4-5}
ChestXR~\cite{ChestXR} & Pneumonia classification & & 17,955 & 3,430\\ \hline
ISIC~\cite{ISIC} & Skin lesion segmentation & 2D dermoscopy & 2000 & 600\\ \hline
\end{tabular}
\end{center}
\vspace{-0.4cm}
\end{table}
\subsection{Experimental Details}
\noindent
\textbf{Pre-training setup.}
We set the size of input 2D patches to $224\times224$ and 3D patches to $16\times96\times96$, aiming to weigh the balance between reserving enough information for SSL and reducing computational and spatial complexity to an affordable level. We applied a rich set of data augmentations to create positive views, including colour jittering, Gaussian blur/noise, random crop, zooming, and flip to the inputs for producing two views.
Following~\cite{DINO}, we adopted the AdamW optimizer~\cite{adamw} with a cosine decaying learning rate~\cite{cosine_LR}, a warm-up period of 10 epochs, to train our UniMiSS. We empirically set the initial learning rate to 0.0008, batch size to 192, maximum epochs to 200,
rate parameter $\omega$ to 0.9, and temperature parameter $\tau_t$ and $\tau_s$ to 0.04 and 0.1, respectively.
It took about 2.5 days to pre-train the UniMiSS using 8 NVIDIA V100 GPUs. We understand this is a big GPU consumption but it saves large amount of time and money to collect 3D medical image data, as we use easily-collected 2D data as the fuel.
\noindent
\textbf{Downstream training setup.}
For the classification, we extracted the pre-trained MiT encoder and appended a FC layer with the output channel as the number of classes for prediction.
For the segmentation, we took the pre-trained MiT encoder and decoder while removing the SSL token, and appended a segmentation head for prediction.
This head includes a transposed convolutional layer, a Conv-IN-LeakyReLU, and a convolutional layer with the kernel size of 1 and the output channel as the number of classes.
The segmentation performance is measured by the Dice coefficient scores. The classification performance is measured by the area under the receiver operator curve (AUC).
Note that we randomly split 25\% training samples as a validation set to select the hyper-parameters of UniMiSS in the ablation study.
The detailed training setups for each downstream task are shown in Appendix.
\begin{table*}[t!]
\caption{Segmentation and classification performance of using different pre-training strategies on the BCV offline test set and RICORD test set.}
\label{tab:tab3}
\vspace{-0.5cm}
\begin{center}
\setlength\tabcolsep{5.5pt}
\begin{tabular}{c|c|l|c|c|c|c|c}
\hline
{\multirow{2}{*}{Methods}} & {\multirow{2}{*}{Backbone}} & \multicolumn{3}{c|}{BCV (CT, seg)} & \multicolumn{3}{c}{RICORD (CT, cls)} \\ \cline{3-8}
& & 20\% & 40\% & 100\% & 20\% & 40\% & 100\% \\ \hline
Rand. init. & \multirow{4}{*}{CNN} & 68.44 & 73.14 & 79.93 & 69.72 & 74.66 & 83.36 \\ \cline{1-1} \cline{3-8}
MoCo v2~\cite{MoCov2} & & 71.22\iffalse\scriptsize{{\color{Iris}$\uparrow$2.78}}\fi & 75.09\iffalse\scriptsize{{\color{Iris}$\uparrow$1.95}}\fi & 82.05\iffalse\scriptsize{{\color{Iris}$\uparrow$2.12}}\fi & 73.46\iffalse\scriptsize{{\color{Iris}$\uparrow$3.74}}\fi & 77.81\iffalse\scriptsize{{\color{Iris}$\uparrow$3.15}}\fi & 85.46\iffalse\scriptsize{{\color{Iris}$\uparrow$2.10}}\fi \\ \cline{1-1} \cline{3-8}
PGL~\cite{PGL} & & 72.05\iffalse\scriptsize{{\color{Iris}$\uparrow$3.61}}\fi & 75.86\iffalse\scriptsize{{\color{Iris}$\uparrow$2.72}}\fi & 82.57\iffalse\scriptsize{{\color{Iris}$\uparrow$2.64}}\fi & 73.76\iffalse\scriptsize{{\color{Iris}$\uparrow$4.04}}\fi & 77.96\iffalse\scriptsize{{\color{Iris}$\uparrow$3.30}}\fi & 85.61\iffalse\scriptsize{{\color{Iris}$\uparrow$2.25}}\fi \\ \cline{1-1} \cline{3-8}
PCRL~\cite{PCRL} & & 72.80\iffalse\scriptsize{{\color{Iris}$\uparrow$4.36}}\fi & 76.05\iffalse\scriptsize{{\color{Iris}$\uparrow$2.91}}\fi & 82.73\iffalse\scriptsize{{\color{Iris}$\uparrow$2.80}}\fi & 75.11\iffalse\scriptsize{{\color{Iris}$\uparrow$5.39}}\fi & 79.01\iffalse\scriptsize{{\color{Iris}$\uparrow$4.35}}\fi & 86.21\iffalse\scriptsize{{\color{Iris}$\uparrow$2.85}}\fi \\ \hline
Rand. init. & \multirow{4}{*}{Transformer} & 70.09 & 74.60 & 79.97 & 71.36 & 76.06 & 83.21 \\ \cline{1-1} \cline{3-8}
MoCo v3~\cite{MoCov3} & & 74.54\iffalse\scriptsize{{\color{Iris}$\uparrow$4.45}}\fi & 78.16\iffalse\scriptsize{{\color{Iris}$\uparrow$3.56}}\fi & 82.02\iffalse\scriptsize{{\color{Iris}$\uparrow$2.05}}\fi & 75.56\iffalse\scriptsize{{\color{Iris}$\uparrow$4.20}}\fi & 79.66\iffalse\scriptsize{{\color{Iris}$\uparrow$3.60}}\fi & 85.16\iffalse\scriptsize{{\color{Iris}$\uparrow$1.95}}\fi \\ \cline{1-1} \cline{3-8}
DINO~\cite{DINO} & & 75.33\iffalse\scriptsize{{\color{Iris}$\uparrow$5.24}}\fi & 78.88\iffalse\scriptsize{{\color{Iris}$\uparrow$4.28}}\fi & 82.61\iffalse\scriptsize{{\color{Iris}$\uparrow$2.64}}\fi & 76.31\iffalse\scriptsize{{\color{Iris}$\uparrow$4.95}}\fi & 80.11\iffalse\scriptsize{{\color{Iris}$\uparrow$4.05}}\fi & 85.91\iffalse\scriptsize{{\color{Iris}$\uparrow$2.70}}\fi \\ \cline{1-1} \cline{3-8}
UniMiSS (Ours) & & \textbf{77.96}\iffalse\scriptsize{{\color{Iris}$\uparrow$7.87}}\fi & \textbf{80.97}\iffalse\scriptsize{{\color{Iris}$\uparrow$6.37}}\fi & \textbf{84.99}\iffalse\scriptsize{{\color{Iris}$\uparrow$5.02}}\fi & \textbf{78.71}\iffalse\scriptsize{{\color{Iris}$\uparrow$7.35}}\fi & \textbf{82.96}\iffalse\scriptsize{{\color{Iris}$\uparrow$6.90}}\fi & \textbf{89.06}\iffalse\scriptsize{{\color{Iris}$\uparrow$5.85}}\fi \\ \hline
\end{tabular}
\end{center}
\vspace{-0.4cm}
\end{table*}
\subsection{Results on 3D downstream tasks}
\noindent
\textbf{Dimension-specific SSL~\textit{vs.} Cross-dimension SSL.}
In this section, we evaluate the SSL performance on two downstream 3D taks, \ie, multi-organ segmentation (BCV) and COVID-19 screening (RICORD).
The UniMiSS pre-training is compared with the random initialization (Rand. init.) and five advanced SSL methods, including MoCo v2/v3~\cite{MoCov2,MoCov3}, PGL~\cite{PGL}, PCRL~\cite{PCRL}, and DINO~\cite{DINO}.
Note that MoCo v2, PGL, and PCRL take the CNN as their encoder backbone, \ie, a 3D ResNet with 50 learnable layers. During the SSL process, MoCo v2 and PGL only pre-train the encoder part, while PCRL additionally pre-trains a decoder by using the reconstruction task.
Besides, MoCo v3, DINO, and our UniMiSS use the Transformer model as the backbone, which contains both encoder and decoder.
We employ the U-like PVT as the backbone for MoCo v3 and DINO, which has a similar architecture of MiT but the different patch embedding module. The lack of SPE make them fail to process both 2D and 3D images simultaneously, resulting in the dimension-specific SSL with only 3D data.
For a fair comparison, all of these SSL methods are pre-trained on the 5,022 unlabeled 3D CT scans. Somewhat differently, the proposed UniMiSS introduces the additional 2D X-rays to the 3D SSL training, benefiting from the universality.
We make more detailed comparisons between the proposed UniMiSS and other dimension-specific CNN/Transformer SSL methods.
Table~\ref{tab:tab3} shows the results of three label settings (20\%,40\%, and 100\% label available).
We summarize this table in the following points:
(1) The Transformer-based models outperform obviously the CNN-based methods, mainly owing to the SSL pre-training.
It reflects that the Transformer is a competitive architecture and the SSL pre-training is essential for the Transformer to achieve good performance.
(2) The proposed UniMiSS is superior to MoCo v3 and DINO. The performance gains over DINO are +2.38\% for segmentation and +3.15\% for classification when 100\% labels are available.
It proves the effectiveness of using a wealth of 2D medical images to assist the 3D SSL process.
(3) Besides, it is really encouraging to see that the proposed UniMiSS is able to achieve the comparable or even superior performance while less annotations, even a half. Taking BCV for example, UniMiSS with 40\% label achieves 80.97\% segmentation Dice, which is better than the 79.97\% of the random initialized method with 100\% labels.
\noindent
\textbf{Comparisons on the BCV online test set. }
To be more persuasive, we also compared the proposed UniMiSS with other state-of-the-art segmentation methods on the BCV online test set. As listed in Table~\ref{tab:tab_on}, these compared methods include PaNN~\cite{PaNN}, UNETR~\cite{Unetr}, nnUnet~\cite{nnUnet} and DoDnet~\cite{dodnet}.
Note that the performance records of these competitors come from their original paper.
It reveals that our UniMiSS, without using any ensemble strategy, still achieves the competitive performance, the best Hausdorff distance (HD) and average mean surface distance (SD), and second highest Dice on the online test set, outperforming the DoDNet with supervised pre-training. When using the coarse-to-fine ensemble strategy like~\cite{nnUnet}, our UniMiSS can obtain the best performance in terms of all metrics.
\begin{table}[t]
\caption{Comparisons on the BCV online test set. }
\label{tab:tab_on}
\vspace{-0.8cm}
\begin{center}
\setlength\tabcolsep{3.5pt}
\begin{tabular}{c|c|c|c|c|c|c}
\hline
Metrics & PaNN~\cite{PaNN} & UNETR~\cite{Unetr} & nnUnet~\cite{nnUnet} & DoDnet~\cite{dodnet} & UniMiSS & UniMiSS \\ \hline
Ensemble & 5 & 5 & 10 & 5 & 1 & 10 \\ \hline
Dice & 85.00 & 85.55 & 87.62 & 86.44 & 87.05 & ~\textbf{88.11} \\ \hline
HD & 18.47 & \textbackslash{} & \textbackslash{} & 15.62 & 13.92 & ~\textbf{13.17} \\ \hline
SD & 1.45 & \textbackslash{} & \textbackslash{} & 1.17 & 1.02 & ~\textbf{0.90} \\ \hline
\end{tabular}
\end{center}
\vspace{-0.3cm}
\end{table}
\noindent
\textbf{Results on 2D downstream tasks.}
Since pre-training on both 2D and 3D medical images, our UniMiSS can be freely applied to 2D downstream tasks.
Table~\ref{tab:tab2} makes the comparisons on the 2D medical image segmentation and classification tasks.
The compared methods include the Rand. init., ImageNet pre-training (INpre), CNN-based SSL methods (\ie MoCo v2, PGL and PCRL), and Transformer-based SSL methods (\ie MoCo v3 and DINO).
Different from the 3D scenarios, a 2D ResNet-50 is used as the backbone in MoCo v2, PGL, and PCRL.
MoCo v3, and DINO still take the U-like PVT as the backbone, but modify the patch embedding to adapt for the 2D inputs.
Here, all compared SSL methods are pre-trained on the same 2D unlabeled medical images.
As for UniMiSS, we directly apply the previous pre-trained model to the 2D tasks, without any modification or further re-training.
From the results, we can find that
(1) the SSL methods have surpassed INpre in both tasks, revealing that pre-training on a large-scale medical image dataset is more friendly to medical domain downstream tasks than pre-training on natural images;
(2) although the number of 3D data is much smaller than 2D, \ie, about one in twenty, the UniMiSS pre-training still achieves the performance gain over the pure 2D SSL method, like DINO.
This may account in part for the inherent correlation between X-rays and CTs. Such a correlation information can be captured by the UniMiSS, thus contributed for the performance gain.
\begin{table*}[t!]
\caption{Segmentation and classification performance of using different pre-training strategies on two 2D test sets.}
\label{tab:tab2}
\vspace{-0.7cm}
\begin{center}
\setlength\tabcolsep{6pt}
\begin{tabular}{c|c|l|l|l|l|l|l}
\hline
{\multirow{2}{*}{Methods}} & {\multirow{2}{*}{Backbone}} & \multicolumn{3}{c|}{JSRT (X-ray, seg)} & \multicolumn{3}{c}{ChestXR (X-ray, cls)} \\ \cline{3-8}
& & 20\% & 40\% & 100\% & 20\% & 40\% & 100\% \\ \hline
Rand. init. & \multirow{5}{*}{CNN} & 84.05 & 87.63 & 90.96 & 92.05 & 94.83 & 97.54 \\ \cline{1-1} \cline{3-8}
INpre~\cite{IN_pretrain} & & 87.90\iffalse\scriptsize{{\color{Iris}$\uparrow$3.85}}\fi & 90.01\iffalse\scriptsize{{\color{Iris}$\uparrow$2.38}}\fi & 91.73\iffalse\scriptsize{{\color{Iris}$\uparrow$0.77}}\fi & 94.78\iffalse\scriptsize{{\color{Iris}$\uparrow$2.73}}\fi & 96.26\iffalse\scriptsize{{\color{Iris}$\uparrow$1.43}}\fi & 98.13\iffalse\scriptsize{{\color{Iris}$\uparrow$0.59}}\fi \\ \cline{1-1} \cline{3-8}
MoCo v2~\cite{MoCov2} & & 88.65\iffalse\scriptsize{{\color{Iris}$\uparrow$4.60}}\fi & 91.03\iffalse\scriptsize{{\color{Iris}$\uparrow$3.40}}\fi & 92.32\iffalse\scriptsize{{\color{Iris}$\uparrow$1.36}}\fi & 95.22\iffalse\scriptsize{{\color{Iris}$\uparrow$3.17}}\fi & 96.61\iffalse\scriptsize{{\color{Iris}$\uparrow$1.78}}\fi & 98.67\iffalse\scriptsize{{\color{Iris}$\uparrow$1.13}}\fi \\ \cline{1-1} \cline{3-8}
PGL~\cite{PGL} & & 89.01\iffalse\scriptsize{{\color{Iris}$\uparrow$4.96}}\fi & 91.39\iffalse\scriptsize{{\color{Iris}$\uparrow$3.76}}\fi & 92.76\iffalse\scriptsize{{\color{Iris}$\uparrow$1.80}}\fi & 95.56\iffalse\scriptsize{{\color{Iris}$\uparrow$3.51}}\fi & 96.96\iffalse\scriptsize{{\color{Iris}$\uparrow$2.13}}\fi & 98.87\iffalse\scriptsize{{\color{Iris}$\uparrow$1.33}}\fi \\ \cline{1-1} \cline{3-8}
PCRL~\cite{PCRL} & & 89.55\iffalse\scriptsize{{\color{Iris}$\uparrow$5.50}}\fi & 91.53\iffalse\scriptsize{{\color{Iris}$\uparrow$3.90}}\fi & 93.07\iffalse\scriptsize{{\color{Iris}$\uparrow$2.11}}\fi & 95.88\iffalse\scriptsize{{\color{Iris}$\uparrow$3.83}}\fi & 97.43\iffalse\scriptsize{{\color{Iris}$\uparrow$2.60}}\fi & 98.99\iffalse\scriptsize{{\color{Iris}$\uparrow$1.45}}\fi \\ \hline
Rand. init. & \multirow{4}{*}{Transformer} & 85.55 & 88.83 & 91.22 & 92.80 & 95.20 & 97.04 \\ \cline{1-1} \cline{3-8}
MoCo v3~\cite{MoCov3} & & 90.07\iffalse\scriptsize{{\color{Iris}$\uparrow$4.52}}\fi & 91.75\iffalse\scriptsize{{\color{Iris}$\uparrow$2.92}}\fi & 92.68\iffalse\scriptsize{{\color{Iris}$\uparrow$1.46}}\fi & 95.99\iffalse\scriptsize{{\color{Iris}$\uparrow$3.19}}\fi & 97.33\iffalse\scriptsize{{\color{Iris}$\uparrow$2.13}}\fi & 98.59\iffalse\scriptsize{{\color{Iris}$\uparrow$1.55}}\fi \\ \cline{1-1} \cline{3-8}
DINO~\cite{DINO} & & 90.40\iffalse\scriptsize{{\color{Iris}$\uparrow$4.85}}\fi & 92.16\iffalse\scriptsize{{\color{Iris}$\uparrow$3.34}}\fi & 93.03\iffalse\scriptsize{{\color{Iris}$\uparrow$1.81}}\fi & 96.44\iffalse\scriptsize{{\color{Iris}$\uparrow$3.64}}\fi & 97.69\iffalse\scriptsize{{\color{Iris}$\uparrow$2.49}}\fi & 98.70\iffalse\scriptsize{{\color{Iris}$\uparrow$1.66}}\fi \\ \cline{1-1} \cline{3-8}
UniMiSS & & \textbf{91.88}\iffalse\scriptsize{{\color{Iris}$\uparrow$6.33}}\fi & \textbf{93.15}\iffalse\scriptsize{{\color{Iris}$\uparrow$4.32}}\fi & \textbf{94.08}\iffalse\scriptsize{{\color{Iris}$\uparrow$2.86}}\fi & \textbf{97.09}\iffalse\scriptsize{{\color{Iris}$\uparrow$4.29}}\fi & \textbf{98.14}\iffalse\scriptsize{{\color{Iris}$\uparrow$2.94}}\fi & \textbf{99.07}\iffalse\scriptsize{{\color{Iris}$\uparrow$2.03}}\fi \\ \hline
\end{tabular}
\end{center}
\vspace{-0.7cm}
\end{table*}
\begin{figure*}[t!]
\begin{center}
{\includegraphics[width=1.0\linewidth]{figs/tiny_vs_small.pdf}}
\end{center}
\vspace{-0.5cm}
\caption{Results of MiT with fewer Transformer layers. Here, MiT-7 and MiT-22 denote MiT with 7 and 22 Transformer layers, respectively.
{\color{gray}\rule[-0.1em]{0.8em}{0.8em}} Rand. init., {\color{Denim}\rule[-0.1em]{0.8em}{0.8em}} Dimension-specific pre-training, {\color{Green(ryb)}\rule[-0.1em]{0.8em}{0.8em}} UniMiSS. Note that the performance gain with yellow and orange color is computed by comparing to the Rand. init., and dimension-specific pre-training baseline, respectively. }
\label{fig:tiny_vs_small}
\end{figure*}
\subsection{Discussions}
\noindent
\textbf{Effectiveness of volume-slice consistency.}
We design the volume-slice consistency mechanism for learning rich representations with 3D medical images. To evaluate the effectiveness of this mechanism, we pre-trained UniMiSS on 3D medical images with or without using the volume-slice consistency.
Table~\ref{tab:tab5} gives the downstream performance on the validation of two 3D datasets. The proposed volume-slice consistency can substantially improve the 3D segmentation/classification accuracy under different label ratios. The performance gain is at least by 1.42\% on segmentation and by 1.47\% on classification.
\begin{table}[t!]
\caption{Segmentation and classification performance on two 3D validation sets with or without using volume-slice consistency.}
\label{tab:tab5}
\vspace{-0.5cm}
\begin{center}
\setlength\tabcolsep{6pt}
\begin{tabular}{c|c|c|c|c|c|c|c}
\hline
\multicolumn{2}{c|}{Objective for 3D} & \multicolumn{3}{c|}{BCV (seg)} & \multicolumn{3}{c}{RICORD (cls)} \\ \hline
Volume & Slices & 20\% & 40\% & 100\% & 20\% & 40\% & 100\% \\ \hline
\checkmark & & 72.08 & 76.04 & 80.94 & 69.87 & 74.61 & 80.96 \\ \hline
\checkmark & \checkmark & \textbf{74.56} & \textbf{77.97} & \textbf{82.36} & \textbf{72.46} & \textbf{76.89} & \textbf{82.43} \\ \hline
\end{tabular}
\end{center}
\vspace{-0.6cm}
\end{table}
\noindent
\textbf{Number of iteration interval.}
The UniMiSS is optimized in a 2D-3D alternation training way, where the iteration interval $\upsilon$ is a critical parameter.
A smaller $\upsilon$ may lead to insufficient training for each domain.
A larger $\upsilon$ may make the network forget the information learned from another domain.
To set a suitable $\upsilon$, we pre-trained UniMiSS with various of $\upsilon$, varying from 1 to 3, and fine-tuned them on four downstream tasks.
Table~\ref{tab:tab6} shows that the pre-trained UniMiSS can achieve the best performance on four downstream tasks when $\upsilon$ equals $2$, and below or above $2$ gives rise to the performance loss.
Hence, we suggest setting the iteration interval to 2 during the cross-domain pre-training.
\begin{table}[t]
\caption{Segmentation and classification performance of our UniMiSS with different iteration intervals on the validation sets.}
\label{tab:tab6}
\vspace{-0.8cm}
\begin{center}
\begin{tabular}{c|c|c|c|c}
\hline
Iteration interval & BCV (3D seg) & RICORD (3D cls) & JSRT (2D seg) & ChestXR (2D cls) \\ \hline
1 & 82.70 & 82.95 & 92.33 & 96.65 \\ \hline
2 & \textbf{83.56} & \textbf{84.26} & \textbf{93.48} & \textbf{97.57} \\ \hline
3 & 83.28 & 83.65 & 93.12 & 97.16 \\ \hline
\end{tabular}
\end{center}
\vspace{-0.3cm}
\end{table}
\noindent
\textbf{MiT with different Transformer scales.}
Transformer is the dominant component in the MiT backbone.
We investigate the effect of Transformer scales in MiT. Specifically, we compare a MiT with 22 Transformer layers (MiT-22) and another with seven layers (MiT-7).
The segmentation and classification performance is given in Figure~\ref{fig:tiny_vs_small}, from which three conclusions can be drawn:
(1) increasing the Transformer layers boosts the performance of MiT in all downstream tasks;
(2) as MiT goes deeper, the performance gain of the dimension-specific pre-training over the random initialization becomes smaller, while the performance gain of our UniMiSS with cross-dimension pre-training is basically impregnable; and
(3) the superiority of our UniMiSS pre-training over the dimension-specific pre-training is more evident with the increase of Transformer layers.
\noindent
\textbf{Transferability on unseen modality data.}
In the above experiments, the pre-training and downstream tasks are all based on CT and X-ray images. To evaluate the transferability of UniMiSS on unseen modalities, we further tested the MoCo v3, DINO and our UniMiSS on the CHAOS dataset (MRI scans) and ISIC dataset (dermoscopic images).
The results in Table~\ref{tab:tab7} show that UniMiSS can consistently
improve at least 2.98\% on the CHAOS dataset, and 1.60\% on the ISIC dataset, compared to the random initialization.
It demonstrates that UniMiSS has a great potential in transferring learned knowledge to the unseen modality. Besides, our UniMiSS also outperforms two popular Transformer-based SSL methods on both CHAOS and ISIC datasets.
\begin{table}[t!]
\caption{Segmentation performance of using the random initialization and three pre-training strategies on CHAOS dataset (unseen MRI scans) and ISIC dataset (unseen dermoscopic images).}
\label{tab:tab7}
\vspace{-0.5cm}
\begin{center}
\setlength\tabcolsep{6pt}
\begin{tabular}{c|c|c|c|c|c|c}
\hline
\multirow{3}{*}{Methods} & \multicolumn{6}{c}{Downstream data} \\ \cline{2-7}
& \multicolumn{3}{c|}{2D dermoscopic} & \multicolumn{3}{c}{3D MRI} \\ \cline{2-7}
& 20\% & 40\% & 100\% & 20\% & 40\% & 100\% \\ \hline
Rand. init. & 76.31 & 79.92 & 85.07 & 73.28 & 83.64 & 88.38 \\ \hline
MoCo v3 & 78.66 & 81.46 & 86.04 & 78.42 & 87.22 & 89.83 \\ \hline
DINO & 79.11 & 81.89 & 86.21 & 79.16 & 87.79 & 90.52 \\ \hline
UniMiSS & \textbf{79.78} & \textbf{82.33} & \textbf{86.67} & \textbf{80.50} & \textbf{88.58} & \textbf{91.36} \\ \hline
\end{tabular}
\end{center}
\vspace{-0.4cm}
\end{table}
\noindent
\textbf{Necessity of SPE.}
Without the SPE module, a straightforward solution is to flatten the pixels or patches and then use a linear layer for the embedding.
Such a crude flattening operation suffers the high computation complexity and memory requirements, especially for 3D images.
Accordingly, SPE is an indispensable part of UniMiSS, which enables to
(1) adaptively choose the patch embedding according to the input type; and
(2) lessen the length of the sequence to reduce computation cost when the network goes deep.
\noindent
\textbf{Visualization of Segmentation Results.}
In Figure~\ref{fig:fig4}, we visualize the segmentation results obtained by the segmentation network, which is initialized (1) randomly, (2) by using the pre-trained MoCo v3~\cite{MoCov3}, (3) by using the pre-trained DINO~\cite{DINO}, or (4) by using our pre-trained UniMiSS.
It shows that our UniMiSS pre-training produces the higher-quality segmentation results, which are more similar to the ground truth, than MoCo v3 and DINO pre-training. Compared to other competitors, UniMiSS pre-training is superior to process challenging cases, like small objects or blurry boundaries.
\begin{figure*}[t!]
\begin{center}
{\includegraphics[width=0.98\linewidth]{figs/Figs_appendix4.pdf}}
\end{center}
\vspace{-0.6cm}
\caption{Visualization of segmentation results of 8 cases selected from four datasets. The regions in red rectangles indicate our superiority. Our UniMiSS pre-training results in more accurate results than random initialization and other two pre-training strategies. Each type of organs and tumors in single dataset is denoted by a unique color.}
\label{fig:fig4}
\vspace{-0.3cm}
\end{figure*}
\section{Conclusion}
We propose a simple yet effective UniMiSS framework, which introduces a wealth of 2D medical images (\ie X-rays) to the 3D SSL, aiming at making up for the lack of 3D data (\ie CT scans).
To break the difficulty of dimensionality barrier, we design the MiT as a bridge to connect different dimensions.
In the future, we will extend our UniMiSS to deal with more dimensions (\eg clinic text or genetic data).
\noindent
\textbf{Acknowledgement}
Jianpeng Zhang and Yong Xia were supported by National Natural Science Foundation of China under Grants 62171377.
Qi Wu was funded by ARC DE190100539.
\bibliographystyle{splncs04}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 7,851
|
\section{Introduction}
\subsection{Some Notation}
Let $\alpha=[a_{0};a_{1},\ldots]$ denote the continued fraction expansion of $\alpha\in\R\setminus\Z$.
We write $||x||$ for the distance from $x$ to the nearest integer.
The \textit{convergents} $p_{n}/q_{n}=[a_{0};a_{1},\ldots, a_{n}]$, where $(p_{n},q_{n})=1$, give good approximations to $\alpha$.
We call $\{q_{n}\}_{n\in\N}$ the \textit{sequence of denominators} of $\alpha$.
We say that an irrational number $\alpha$ is \textit{badly-approximable} if there exists $\varepsilon_{\alpha}>0$ such that for all $p,q\in\Z$, $(p,q)=1$, we have
\begin{equation}
\left|\alpha-\frac{p}{q}\right|>\frac{\varepsilon_{\alpha}}{q^{2}}.
\end{equation}
These correspond precisely with those numbers $\alpha$ for which there exists $N\in\N$ such that $a_{n}(\alpha)\leq N$ for all $n\in\N$. The set of all badly-approximable numbers is a set of Lebesgue measure zero.\\
When $\alpha$ is badly approximable, we have the helpful bound that
\begin{equation}
||q_{n}\alpha||>\frac{\varepsilon_{\alpha}}{q_{n}}
\end{equation}
and since convergents give the best approximations for the distance to the nearest integer (see \cite{Khin}), this means that for $m\leq q_{n+1}-1$ we have the bound
\begin{equation}
||m\alpha||>\frac{\varepsilon_{\alpha}}{q_{n}}.
\end{equation}
We write $f(n)=O(g(n))$ to mean that there exists a constant $C$ (which doesn't depend on $n$), such that $f(n)\leq C\cdot g(n)$ for all $n\in\N$.\\
Finally we define the discrepancy of a sequence.
\begin{defn}
\label{discrep}
Let $(x_{n})$ be a sequence of real numbers. For $N\in\N$ the \textit{discrepancy} of $(x_{n})$ modulo one, $D_{N}(x_{n})$, is defined as:
\begin{equation}
D_{N}(\{x_{m}\}):=\sup_{I\subseteq\mathbb{R}/\mathbb{Z}}\left|\sum_{m=1}^{N}\chi_{I}(x_{m})-N\cdot|I|\right|
\end{equation}
where $I$ denotes an interval and $\chi_{I}$ is the characteristic function of $I$.
\end{defn}
\subsection{Double Exponential Sums}
In \cite{Corinna} Sinai \& Ulcgrai studied double trigonometric sums of the form
\begin{equation}
\label{equation}
T_{M}(\alpha)=\frac{1}{M}\sum_{m=0}^{M-1}\sum_{n=0}^{M-1}{e(nm\alpha)}.
\end{equation}
We want to determine when the absolute value of this sum is bounded uniformly (i.e. by a constant which depends only on $\alpha$) over some subsequence $M\in\mathscr{A}\subseteq\N$. This will obviously depend on the Diophantine properties of $\alpha$ and the subsequence $\mathscr{A}$.\\
We will see that the problem of bounding this sum depends importantly on controlling sums such as
\begin{equation}
\label{simple}
\left|\frac{1}{M}\sum_{m=1}^{M}\frac{1}{\{\{m\alpha\}\}}\right|.
\end{equation}
Here
\begin{equation}
\{\{x\}\}:=
\begin{cases}
\{x\}, & x\in [0,\frac{1}{2}],\\
\{x\}-1, & x\in (-\frac{1}{2},0),
\end{cases}
\end{equation}
is the \textit{signed fractional part} of $x\in\R$.\\
\\
In \cite{Corinna} the following is proved
\begin{thm}[Sinai, Ulcigrai]
\label{thm1}
Let $\alpha$ be badly-approximable. Consider the following double trigonometric sum:
\begin{equation}
\label{goal}
T_{M}(\alpha):=\frac{1}{M}\sum_{m=0}^{M-1}\sum_{n=0}^{M-1}e(nm\alpha).
\end{equation}
Then there exists a constant $C=C(\alpha)>0$ such that $|T_{M}|\leq C_{\alpha}$ for all $M\in\{q_{n}\}_{n\in\N}$.
\end{thm}
Our main theorem generalises this.
\begin{thm}
\label{mainthm}
Let $\alpha\in\R\setminus\Z$. Then there exists a constant $C=C(\alpha)>0$ such that
\begin{equation}
|T_{q_{n}}|\leq C_{\alpha}\cdot\max\left\{\frac{\log (2\cdot\max_{i\leq n}\{a_{i}\})}{a_{n+1}},1\right\}.
\end{equation}
for all $n\in\N$.
\end{thm}
\begin{rmk}
By examining signs it appears that the upper bound here is close to best possible. Equation (1.13) in \cite{BHV} gives a lower bound for the largest terms in a sum that we will consider. While it's true that we use an inequality earlier in the calculation, it isn't too restrictive.
\end{rmk}
\section{Proof of main result}
\subsection{Reducing $T_{M}$}
Following the methods in \cite{Corinna} we split $T_{M}$ into two separate sums.\\
By summing the terms for $n=0,\ldots, M-1$ we can rewrite (\ref{equation}) as
\begin{equation}
T_{M}=1+\frac{1}{M}\sum_{m=1}^{M-1}{\frac{e(Mm\alpha)-1}{e(m\alpha)-1}}.
\end{equation}
Then we can write $T_{M}=1+S'_{M}-S''_{M}$ where
\begin{align}
S'_{M}:=&\frac{1}{M}\sum_{m=1}^{M-1}{\frac{e(Mm\alpha)}{e(m\alpha)-1}}\ \text{and}\label{sum1}\\
S''_{M}:=&\frac{1}{M}\sum_{m=1}^{M-1}{\frac{1}{e(m\alpha)-1}}.\label{sum2}
\end{align}
We will prove that there exist constants $C',C''\in\R$ such that
\begin{equation}
\label{anewstep}
|S'_{q_{n}}|\leq C'\cdot\max\left\{\frac{\log (2\cdot\max_{i\leq n}\{a_{i}\})}{a_{n+1}},1\right\}
\end{equation}
and
\begin{equation}
|S''_{q_{n}}|\leq C'',
\end{equation}
for all $n\in\N$. These constants will depend only on $\alpha$.
\subsection{The Sum $S''_{M}$ (\ref{sum2})}
Let's consider the `less intimidating' sum first.
We want to show that there exists $C''$ such that $|S''_{q_{n}}|\leq C''$ for all $n\in\N$.\\
Note that in \cite{HL}, Hardy and Littlewood prove a similar theorem.
\begin{thm}[Hardy, Littlewood]
\label{HL}
Let $\alpha$ be badly-approximable. Then there exists $C^{*}>0$ such that $|S^{''}_{M}|\leq C^{*}$ for each $M\in\N^{+}$.
\end{thm}
We proceed by calculating real and imaginary parts.
\begin{equation}
\frac{1}{e(m \alpha)-1}=-\frac{1}{2}-\frac{i}{2} \cot(\pi m \alpha).
\end{equation}
The Taylor series expansion of $\cot x$ is
\begin{equation}
\cot x= \sum_{n=0}^{\infty}\frac{(-1)^{n}4^{n}B_{2n}}{(2n)!}x^{2n-1}=\frac{1}{x}-\frac{x}{3}+\frac{x^{3}}{45}-\ldots
\end{equation}
with radius of convergence $0<|x|<\pi$. Here $B_{n}$ is the $n$th \textit{Bernoulli number}.\\
Note that due to the symmetry of $\cot x$,
\begin{equation}\cot(\pi m\alpha)=\cot(\pi \{\{m\alpha\}\}).
\end{equation}
So we can write
\begin{equation}
\cot(\pi m\alpha)=\frac{1}{\pi\{\{m\alpha\}\}}\left(1+\sum_{n=1}^{\infty}\frac{(-1)^{n}4^{n}B_{2n}}{(2n)!}(\pi\{\{m\alpha\}\})^{2n}\right)
\end{equation}
Now the series on the right is negative and it takes values strictly between $0$ (when $\{\{m\alpha\}\} $ is close to $0$) and $-1$ (when $\{\{m\alpha\}\} $ is close to $\pm\frac{1}{2}$).
\\
Hence in order to prove that $|S^{''}_{q_{n}}|$ is bounded by a uniform constant for all $n\in\N$, we have to prove the following
\begin{lem}
\label{new}
Let $\alpha\in\R$. Then there exists $C=C(\alpha)>0$ such that, for all $n\in \N$,
\begin{equation}
\label{stepstone}
\left|\sum_{m=1}^{q_{n}-1}\frac{1}{q_{n}\{\{m\alpha\}\}}\right|\leq C.
\end{equation}
\end{lem}
We will consider two different proofs of Lemma \ref{new}. The first is simpler, while the the latter will be applicable to $S'_{M}$ as well. The second proof is also malleable to proving Theorem \ref{HL}.
\subsection{Koksma-Hlawka Proof of Lemma \ref{new}}
Recall the Koksma-Hlawka inequality.
\begin{lem}
Let $f$ be a function with period $1$ of bounded variation. Then for every sequence $x_{m}$ and every integer $N\geq 1$, we have
\begin{equation*}
\left|\frac{1}{N}\sum_{m=1}^{N}{f(x_{m})}-\int_{0}^{1}f(x)dx\right|\leq V(f)\frac{D_{N}(x_{m})}{N},
\end{equation*}
where $V(f)$ is the total variation of the function.
\end{lem}
We wish to apply this inequality with $f(x)=\frac{1}{\{\{x\}\}}$, $x_{m}=\{m\alpha\}$ and $N=q_{n}-1$.\\
Therefore we have to restrict the domain on which we define our function, in order to ensure that it's integrable.\\
We're able to use the following from \cite{Rock}.
\begin{equation}
\label{useful}
\left|\alpha-\frac{p_{n-1}}{q_{n-1}}\right|>\frac{1}{2q_{n-1}q_{n}}.
\end{equation}
So for all $m\leq N=q_{n}-1$, we have
\begin{equation}
||m\alpha||>\frac{1}{2q_{n}}.
\end{equation}
Hence we can restrict the domain of $f$ to the interval $[\frac{1}{2q_{n}},1-\frac{1}{2q_{n}}]$. Since $f$ is anti-symmetric about $1/2$, the integral above is equal to $0$.\\
The total variation, $V(f)$, of $f$ is
\begin{equation*}
\sup_{p}\sum_{i=1}^{n_{p}}\left|\frac{1}{\{\{x_{i+1}\}\}}-\frac{1}{\{\{x_{i}\}\}}\right|,
\end{equation*}
where $\mathcal{P}$ is a partition of $[\frac{1}{2q_{n}},1-\frac{1}{2q_{n}}]$. It is maximised when we take the trivial partition (that is the two endpoints). Therefore $V(f)=4q_{n}$.\\
Finally we move on to considering the Discrepancy.\\
Lemma $5.6$ from \cite{Harman} states that
\begin{equation}
\label{Harman}
D_{N}(m\alpha)\leq 3\sum_{j=0}^{r}t_{j},
\end{equation}
where $N=\sum_{j=0}^{r}q_{r}t_{j}$. So if $N=q_{n}$ then $t_{n}=1$ and $t_{i}=0$ for all $i\neq n$. So $D_{q_{n}}(m\alpha)\leq 3$.
\\
Finally can apply all the estimates we have (with $N=q_{n}-1$ and $f$ \& $\{x_{m}\}$ as above).
\begin{align}
\left|\sum_{m=1}^{q_{n}-1}f(x_{m})-(q_{n}-1)\int_{0}^{1}f(x)dx\right| & \leq D_{q_{n}-1}(x_{m})V(f)\\
& \leq(D_{q_{n}}(x_{m})+1)V(f)\\
& \leq 4\cdot 4q_{n}=16q_{n}
\end{align}
Hence
\begin{equation}
\left|\frac{1}{q_{n}}\sum_{m=1}^{q_{n}-1}\frac{1}{\{\{m\alpha\}\}}\right|\leq 16.
\end{equation}
Here we used the obvious fact that $D_{M}(x_{m})\leq D_{M+1}(x_{m})+1$.
\subsection{The sum $S^{'}_{M}$ (\ref{sum1})}
We move on to considering the sum
\begin{equation}
S'_{M}:=\frac{1}{M}\sum_{m=1}^{M-1}{\frac{e(Mm\alpha)}{e(m\alpha)-1}},\label{sum3}
\end{equation}
We will write this sum as a telescoping series and take advantage of some cancellation to reduce our situation to considering the sum $S^{''}_{M}$ (\ref{sum2}).\\
\\
We write
\begin{align}
\sum_{m=1}^{M-1}\frac{e(Mm\alpha)}{e(m\alpha)-1}&=\sum_{m=1}^{M-1}(e(Mm\alpha)-e(M(m+1)\alpha))\sum_{k=1}^{m}\frac{1}{e(k\alpha)-1}\label{firstline2}\\ &+e(M^{2}\alpha)\sum_{k=1}^{M-1}\frac{1}{e(k\alpha)-1}\label{basics}.
\end{align}
We write $\alpha=\frac{p_{n}}{q_{n}}+\frac{\xi_{n}}{q_{n}q_{n+1}}$ where $\frac{1}{2}<|\xi_{n}|<1$.
We then consider the outer part of the sum on the right hand side of (\ref{firstline2}) (for $M=q_{n}$)
\begin{align}
e(mq_{n}\alpha)-e((m+1)q_{n}\alpha) & =e(mq_{n}\psi_{n})-e((m+1)q_{n}\psi_{n})\\
& =e(mq_{n}\psi_{n})-e(mq_{n}\psi_{n})e(q_{n}\psi_{n})\\
& =(1-e(q_{n}\psi_{n}))e(mq_{n}\psi_{n}),
\end{align}
which in absolute value is less than $2\pi/q_{n+1}$.\\
Now using the triangle inequality and Lemma \ref{new} we see that (\ref{anewstep}) results from the following lemma
\begin{lem}
\label{ost}
$\forall m\leq q_{n}-1$,
\begin{equation}
\sum_{k=1}^{m}\frac{1}{\{\{k\alpha\}\}}=O(q_{n}\max_{i\leq n}\{1,\log a_{i}\}).
\end{equation}
\end{lem}
To prove this Lemma we will need to introduce some different techniques, which will also yield a new proof of Lemma \ref{new}.
\subsection{Ostrowski Proof of Lemmas \ref{new} \& \ref{ost}}
Our alternative proof of Lemma \ref{new} will be to decompose the sum in (\ref{stepstone}) into segments where there is some obvious cancellation.
\begin{defn}
\label{ostrowski}
Let $\alpha$ be irrational, then for every $n\in\N$ there exists a unique integer $M\geq 0$ and a unique sequence $\{c_{k+1}\}_{k=0}^{\infty}$ of integers such that $q_{M}\leq m<q_{M+1}$ and
\begin{equation}
m=\sum_{k=0}^{\infty}{c_{k+1}q_{k}},
\end{equation}
with
\begin{align}
& 0\leq c_{1}<a_{1},\ 0\leq c_{k+1}\leq a_{k+1}\ \text{for}\ k\geq1,\\
& c_{k}=0\ \text{whenever}\ c_{k+1}=a_{k+1}\ \text{for some}\ k\geq 1,
\end{align}
and
\begin{align}
c_{k+1}=0\ \text{for}\ k>M.
\end{align}
This is known as the \textit{Ostrowski expansion}.
\end{defn}
We will consider segments of our sum which `spread out' in the unit interval. We take our inspiration from a set of intervals discussed in \cite{Haynes}.
\begin{defn}[Special Intervals]
Define $A(m,c)$ to be the collection of non-negative integers $n$ with Ostrowski expansions of the form
\begin{equation}
n=cq_{m-1}+\sum_{k=m}^{\infty}c_{k+1}q_{k}
\end{equation}
and define a set $J(m,c)$ (which turns out be be an interval, see \cite{Haynes}) in $\R/\Z$ by
\begin{equation}
J(m,c)=\overline{\{n\alpha:n\in A(m,c)\}}
\end{equation}
For any fixed $m$, these intervals cover $\R/\Z$ and have some very nice properties such as the discrepancy of $\{n\alpha\}$ being bounded.
We will use what these intervals tell us about the distribution of $n\alpha$ on the unit interval to achieve cancellation in (\ref{simple}).
\end{defn}
Let $m\leq q_{n}-1$,
\begin{equation}
m=\sum_{i=0}^{n-1}c_{i+1}q_{i},\quad 0\leq c_{i+1}\leq a_{i+1}
\end{equation}
and
\begin{equation}
n(i,c):=\sum_{j=0}^{i-1}c_{j+1}q_{j}+cq_{i}.
\end{equation}
We will use this decomposition to sum up to $m$.
\begin{equation}
\label{newish}
\sum_{k=1}^{m}{\frac{1}{\{\{k\alpha\}\}}}=\sum_{i=0}^{n-1}\sum_{c=0}^{c_{i+1}-1}\sum_{l=n(i,c)+1}^{n(i,c)+q_{i}}\frac{1}{\{\{l\alpha\}\}}
\end{equation}
Note that $n(i,c)+q_{i}=n(i,c+1)$ and $n(i,c_{i+1}-1)+q_{i}=n(i+1,0)$.\\
Let's consider a situation where we are studying
\begin{equation}
\label{typical}
\sum_{l=n(i,c)+1}^{n(i,c)+q_{i}}\frac{1}{\{\{l\alpha\}\}}.
\end{equation}
We wish to approximate $\alpha$ by $p_{i}/q_{i}$ and achieve (almost) complete cancellation in the main term that we get.\\
Obviously problems can occur. Specifically, if $l\cdot p_{i}\equiv 0(q_{i})$ then we don't want to divide by $0$, so we want to isolate these terms and deal with them separately. Note that since $(p_{i},q_{i})=1$, we have a complete set of residue classes modulo $q_{i}$, so in each sum, (\ref{typical}), we will have exactly one term, $(c+1)q_{i}$, where this happens .\\
Also, there exists $r\leq q_{i}$ such that
\begin{align}
n(i,0)+r=q_{i}\\
n(i,1)+r=2q_{i}
\end{align}
\begin{equation*}
\vdots
\end{equation*}
\begin{align}
n(i,c_{i+1}-1)+r=c_{i+1}q_{i}
\end{align}
So we can consider all these terms separately.
Finally we consider summing over a complete set of residue classes modulo $q_{i}$. We will first consider the simple case, ($1\leq k\leq q_{i-1}$), which will give us a second proof of Lemma \ref{new}.\\
We write
\begin{equation}
\alpha=\frac{p_{i}}{q_{i}}+\frac{\xi_{i}}{q_{i}q_{i+1}},
\end{equation}
where $\frac{1}{2}<|\xi_{i}|<1$. Now
\begin{align}
\label{thingy}
\sum_{k=1}^{q_{i}-1}{\frac{1}{\{\{k\alpha\}\}}}& =\sum_{k=1}^{q_{i}-1}{\frac{1}{{\{\{k\frac{p_{i}}{q_{i}}+\frac{k\xi_{i}}{q_{i}q_{i+1}}\}\}}}}.
\end{align}
Now we use the fact that
\begin{equation}
\left\{\left\{\frac{kp_{i}}{q_{i}}+\frac{k\xi_{i}}{q_{i}q_{i+1}}\right\}\right\}=\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}+\left\{\left\{\frac{k\xi_{i}}{q_{i}q_{i+1}}\right\}\right\},
\end{equation}
unless perhaps if $kp_{i}\equiv\frac{q_{i}}{2}$ modulo $q_{i}$ (when $2|q_{i}$), or if $kp_{i}\equiv\frac{q_{i}\pm 1}{2}$ (when $2|q_{i}+1$).
Now (\ref{thingy}) equals\footnote{The one or two extra term/s mentioned just above have been removed from the sum and are accounted for by the $O(1)$ term.}
\begin{align}
\sideset{}{'}\sum_{k=1}^{q_{i-1}}{\frac{1}{\{\{k\frac{p_{i}}{q_{i}}\}\}}\left(\frac{1}{1+\{\{\frac{kp_{i}}{q_{i}}\}\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}}\right)}+O(1).
\end{align}
Furthermore
\begin{gather}
\left(\frac{1}{1+\{\{\frac{kp_{i}}{q_{i}}\}\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}}\right)\\
=1-\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}+\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}^{-2}\left(\frac{k\xi_{i}}{q_{i}q_{i+1}}\right)^{2}-\ldots
\end{gather}
There exists $n_{k}$ such that $1\leq n_{k}\leq q_{i}-1$ and $n_{k}\equiv kp_{i}\mod q_{i}$. Now we define $n^{'}_{k}$ as follows
\begin{equation}
n^{'}_{k}:=
\begin{cases}
n_{k}, & n_{k}\leq \frac{q_{i}}{2}\\
n_{k}-q_{i}, & n_{k}>\frac{q_{i}}{2}.
\end{cases}
\end{equation}
Then
\begin{equation}
\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}=\frac{k\xi_{i}}{n^{'}_{k}q_{i+1}}.
\end{equation}
We then know that for all $k$,
\begin{gather}
-\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}+\left\{\left\{\frac{kp_{i}}{q_{i}}\right\}\right\}^{-2}\left(\frac{k\xi_{i}}{q_{i}q_{i+1}}\right)^{2}-\ldots\\
=C_{k}\frac{k\xi_{i}}{n^{'}_{k}q_{i+1}}.
\end{gather}
We need $|n^{'}_{k}|\geq 2$ in order to have a uniform bound over $k$ for the constant $C_{k}$. When this is the case
\begin{equation}
-\frac{1}{2}<C_{k}<2,
\end{equation}
(apart from the one or two exceptions mentioned previously).
So we have to isolate another two terms. We write $k_{1}$, $k_{-1}$ for the numbers where $k_{1}p_{i}\equiv 1\mod q_{i}$ and $k_{-1}p_{i}\equiv -1\mod q_{i}$ respectively.\\
So (\ref{thingy}) becomes
\begin{gather}
\sum_{n_{k}=2}^{q_{i}-2}{\left(\frac{1}{\{\{\frac{n_{k}}{q_{l}}\}\}}+C_{k}\left(\frac{k\xi_{i}q_{i}}{(n^{'}_{k})^{2}q_{i+1}}\right)\right)}+\frac{1}{\{\{k_{1}\alpha\}\}}+\frac{1}{\{\{k_{-1}\alpha\}\}}+O(1)\\
=\sum_{n_{k}=2}^{q_{i}-2}{C_{k}\left(\frac{k\xi_{i}q_{i}}{(n^{'}_{k})^{2}q_{i+1}}\right)}+O(q_{i}).
\end{gather}
(We used the basic approximation from Khinchin (\ref{useful}) to deal with the two extra terms.)\\
By the rearrangement inequality (see \cite{HLP}, Theorem 368) this first sum is less than
\begin{equation}
q_{i}\left(\frac{1}{2^{2}}+\frac{1}{3^{2}}+\ldots\right),
\end{equation}
which in turn is bounded above by $q_{i}$.\\
So
\begin{equation}
\sum_{k=1}^{q_{i}-1}{\frac{1}{\{\{k\alpha\}\}}}=O(q_{i}),
\end{equation}
as required.\\
Now we move on to a proof of Lemma \ref{ost}. We wish to prove (for all $i$) that
\begin{equation}
\sum_{c=0}^{c_{i+1}-1}\sum_{l=n(i,c)+1}^{n(i,c)+q_{i}}\frac{1}{\{\{l\alpha\}\}}=O(q_{i+1}\log c_{i+1}).
\end{equation}
Note that if we sum
\begin{equation}
\sum_{l=n(i,c)+1}^{n(i,c)+q_{i}}\frac{1}{\{\{l\alpha\}\}}
\end{equation}
then a similar argument to the proof of Lemma \ref{new} shows that this is equal to
\begin{equation}
O(q_{i})+\frac{1}{\{\{k_{(1,c)}\alpha\}\}}+\frac{1}{\{\{k_{(-1,c)}\alpha\}\}}+\frac{1}{\{\{(c+1)q_{i}\alpha\}\}},
\end{equation}
where $n(i,c)+1\leq k_{(\pm 1,c)}\leq n(i,c)+q_{i}$, and $k_{(\pm 1,c)}p_{i}\equiv \pm 1\mod q_{i}$.\\
Clearly $k_{(\pm 1,c)}=k_{(\pm 1,0)}+cq_{i}$.\\
Furthermore as $m<q_{i+1}$,
\begin{equation}
k_{(\pm 1,c_{i+1}-r)}<n(i,(c_{i+1}-r))+q_{i}< q_{i+1}-(r-1)q_{i}.
\end{equation}
Now, we calculated earlier that
\begin{equation}
\frac{1}{\{\{k\alpha\}\}}=\frac{1}{\{\{k\frac{p_{i}}{q_{i}}\}\}}\left(\frac{1}{1+\{\{\frac{kp_{i}}{q_{i}}\}\}^{-1}\frac{k\xi_{i}}{q_{i}q_{i+1}}}\right)
\end{equation}
Letting $k=k_{(1,0)}$
\begin{align}
\frac{1}{\{\{k_{(1,0)}\alpha\}\}}&=q_{i}\left(\frac{1}{1+\frac{k_{(1,0)}\xi_{i}}{q_{i+1}}}\right)\\
&= \frac{q_{i}q_{i+1}}{q_{i+1}+k_{(1,0)}\xi_{i}}.
\end{align}
Hence
\begin{align}
\frac{1}{\{\{k_{(1,c)}\alpha\}\}}&=\frac{q_{i}q_{i+1}}{q_{i+1}+(k_{(1,0)}+cq_{i})\xi_{i}}
\end{align}
and also
\begin{align}
\frac{1}{\{\{k_{(-1,c)}\alpha\}\}}&=\frac{-q_{i}q_{i+1}}{q_{i+1}-(k_{(-1,0)}+cq_{i})\xi_{i}}
\end{align}
Now, without loss of generality, assume that $\xi_{i}>0$. Then
\begin{equation}
\frac{1}{\{\{k_{(1,c)}\alpha\}\}}<q_{i}
\end{equation}
for all $c$. Hence
\begin{align}
\sum_{c=0}^{c_{i+1}-1}\sum_{l=n(i,c)+1}^{n(i,c)+q_{i}}{\frac{1}{\{\{l\alpha\}\}}} &=O(q_{i+1})+\sum_{c=0}^{c_{i+1}-1}{\frac{1}{\{\{(c+1)q_{i}\alpha\}\}}}\\&+\sum_{c=0}^{c_{i+1}-1}{\frac{-q_{i}q_{i+1}}{q_{i+1}-(k_{(-1,0)}+cq_{i})\xi_{i}}}\\
&=O(q_{i+1})+O(q_{i+1}\log c_{i+1})\\
&+\sum_{c=0}^{c_{i+1}-1}{\frac{-q_{i}q_{i+1}}{q_{i+1}-(k_{(-1,0)}+cq_{i})\xi_{i}}}.
\end{align}
Finally
\begin{align}
\left|\sum_{c=0}^{c_{i+1}-1}{\frac{-q_{i}q_{i+1}}{q_{i+1}-(k_{(-1,0)}+cq_{i})\xi_{i}}}\right|& \leq\frac{q_{i}}{c_{i+1}-1}+\ldots+\frac{q_{i}}{2}+q_{i}+2q_{i+1}\\
& =O(q_{i+1}\log c_{i+1}).
\end{align}
As this is true for all $i$, the condition for Lemma \ref{ost} follows.
\begin{rmk}
Equation (1.13) in \cite{BHV} tells us that the sum
\begin{equation}
\sum_{c=0}^{c_{i+1}-1}{\frac{1}{\{\{(c+1)q_{i}\alpha\}\}}}
\end{equation}
can be no smaller than $O(q_{i+1}\log c_{i+1})$.
\end{rmk}
\begin{rmk}
In our final calculation we have ignored the cancellation between the positive and negative terms. However, when $c_{i+1}\approx a_{i+1}/2$ for example, we get very little cancellation and our main term is $O(q_{i+1}\log a_{i+1})$
\end{rmk}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 8,384
|
This tutorial explains how to use Access / Excel VBA to delete folder if folder exists already.
Suppose you want to write a Macro to export files to a folder. There could be problem if the folder already contains the files you want to export. To avoid this, you probably want to delete the folder and create a new folder before exporting the files. Alternatively, you can also delete all files in the folder.
In order to delete a folder, you can use the FSO Folder Methods.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,381
|
This is a work of fiction. Names, characters, places, and incidents are either the product of the author's imagination or are used fictitiously, and any resemblance to actual persons, living or dead, business establishments, events or locales is entirely coincidental.
_SILENT PREY_
A _Berkley_ Book / published by arrangement with the author
All rights reserved.
Copyright © _1992_ by _John Sanford_
This book may not be reproduced in whole or part, by mimeograph or any other means, without permission. Making or distributing electronic copies of this book constitutes copyright infringement and could subject the infringer to criminal and civil liability.
For information address:
The Berkley Publishing Group, a division of Penguin Putnam Inc.,
375 Hudson Street, New York, New York 10014.
The Penguin Putnam Inc. World Wide Web site address is
<http://www.penguinputnam.com>
ISBN: _1-101-14624-9_
A _BERKLEY_ BOOK®
_Berkley_ Books first published by The Berkley Publishing Group, a member of Penguin Putnam Inc.,
375 Hudson Street, New York, New York 10014.
_Berkley_ and the " _B_ " design are trademarks belonging to Penguin Putnam Inc.
Electronic edition: May, 2002
## Contents
CHAPTER 1
CHAPTER 2
CHAPTER 3
CHAPTER 4
CHAPTER 5
CHAPTER 6
CHAPTER 7
CHAPTER 8
CHAPTER 9
CHAPTER 10
CHAPTER 11
CHAPTER 12
CHAPTER 13
CHAPTER 14
CHAPTER 15
CHAPTER 16
CHAPTER 17
CHAPTER 18
CHAPTER 19
CHAPTER 20
CHAPTER 21
CHAPTER 22
CHAPTER 23
CHAPTER 24
CHAPTER 25
CHAPTER 26
CHAPTER 27
CHAPTER 28
CHAPTER 29
CHAPTER 30
CHAPTER 31
CHAPTER 32
## CHAPTER
1
A thought sparked in the chaos of Bekker's mind.
_The jury._
He caught it, mentally, like a quick hand snatching a fly from midair.
Bekker slumped at the defense table, the center of the circus. His vacant blue eyes rolled back, pale and wide as a plastic baby-doll's, wandering around the interior of the courtroom, snagging on a light fixture, catching on an electrical outlet, sliding past the staring faces. His hair had been cut jailhouse short, but they had let him keep the wild blond beard. An act of mercy: the beard disguised the tangled mass of pink scar tissue that crisscrossed his face. In the middle of the beard, his pink rosebud lips opened and closed, like an eel's, damp and glistening.
Bekker looked at the thought he'd caught: _The jury._ Housewives, retirees, welfare trash. His _peers,_ they called them. A ridiculous concept: he was a doctor of medicine. He stood at the top of his profession. He was _respected._ Bekker shook his head.
_Understand . . . ?_
The word tumbled from the judge-crow's mouth and echoed in his mind. "Do you understand, Mr. Bekker?"
_What . . . ?_
The idiot flat-faced attorney pulled at Bekker's sleeve: "Stand up."
_What . . . ?_
The prosecutor turned to stare at him, hate in her eyes. The hate touched him, reached him, and he opened his mind and let it flow back. _I'd like to have you for five minutes, good sharp scalpel would open you up like a goddamn oyster: zip, zip. Like a goddamn clam._
The prosecutor felt Bekker's interest. She was a hard woman; she'd put six hundred men and women behind bars. Their petty threats and silly pleas no longer interested her. But she flinched and turned away from Bekker.
_What? Standing? Time now?_
Bekker struggled back. It was so hard. He'd let himself go during the trial. He had no interest in it. Refused to testify. The outcome was fixed, and he had more serious problems to deal with. Like survival in the cages of the Hennepin County Jail, survival without his medicine.
But now the time had come.
His blood still moved too slowly, oozing through his arteries like strawberry jam. He fought, and simultaneously fought to hide his struggle.
_Focus._
And he started, so slowly it was like walking through paste, trudging back to the courtroom. The trial had lasted for twenty-one days, had dominated the papers and the television newscasts. The cameras had ambushed him, morning and night, hitting him in the face with their intolerable lights, the cameramen scuttling backward as they transferred him, in chains, between the jail and the courtroom.
The courtroom was done in blond laminated wood, with the elevated judge's bench at the head of the room, the jury box to the right, tables for the prosecution and defense in front of the judge. Behind the tables, a long rail divided the room in two. Forty uncomfortable spectator's chairs were screwed to the floor behind the rail. The chairs were occupied an hour before arguments began, half of them allotted to the press, the other half given out on a first-come basis. All during the trial, he could hear his name passing through the ranks of spectators: _Bekker Bekker Bekker._
The jury filed out. None of them looked at him. They'd be secluded, his _peers,_ and after chatting for a decent interval, they'd come back and report him guilty of multiple counts of first-degree murder. The verdict was inevitable. When it was in, the crow would put him away.
The black asshole in the next cell had said it, in his phony street dialect: "They gon slam yo' nasty ass into Oak Park, m'man. You live in a motherfuckin' cage the size of a motherfuckin' refrigerator wit a TV watching you every move. You wanta take a shit, they watchin' every move, they makin' movies of it. Nobody ever git outa Oak Park. It is a true motherfucker."
But Bekker wasn't going. The thought set him off again, and he shook, fought to control it.
_Focus_ . . .
He focused on the small parts: The gym shorts biting into the flesh at his waist. The razor head pressed against the back of his balls. The Sox cap, obtained in a trade for cigarettes, tucked under his belt. His feet sweating in the ridiculous running shoes. Running shoes and white socks with his doctor's pinstripes—he looked a fool and he knew it, hated it. Only a moron would wear white socks with pinstripes, but white socks _and_ running shoes . . . no. People would be laughing at him.
He could have worn his wing tips, one last time—a man is innocent until proven guilty—but he refused. They didn't understand that. They thought it was another eccentricity, the plastic shoes with the seven-hundred-dollar suit. They didn't know.
_Focus._
Everyone was standing now, the crow-suit staring, the attorney pulling at his sleeve. And here was Raymond Shaltie . . .
"On your feet," Shaltie said sharply, leaning over him. Shaltie was a sheriff's deputy, an overweight time-server in an ill-fitting gray uniform.
"How long?" Bekker asked the attorney, looking up, struggling to get the words out, his tongue thick in his mouth.
"Shhh . . ."
The judge was talking, looking at them: " . . . standing by, and if you leave your numbers with my office, we'll get in touch as soon as we get word from the jury . . ."
The attorney nodded, looking straight ahead. He wouldn't meet Bekker's eyes. Bekker had no chance. In his heart, the attorney didn't want him to have a chance. Bekker was nuts. Bekker _needed_ prison. Prison forever and several days more.
"How long?" Bekker asked again. The judge had disappeared into her chambers. _Like to get her, too._
"Can't tell. They'll have to consider the separate counts," the attorney said. He was court-appointed, needed the money. "We'll come get you . . ."
_Pig's eye, they would._
"Let's go," said Shaltie. He took Bekker's elbow, dug his fingertips into the nexus of nerves above Bekker's elbow, an old jailer's trick to establish dominance. Unknowingly, Shaltie did Bekker a favor. With the sudden sharp pulse of pain, Bekker snapped all the way back, quick and hard, like a handclap.
His eyes flicked once around the room, his mind cold, its usual chaos squeezed into a high-pressure corner, wild thoughts raging like rats in a cage. Calculating. He put pain in his voice, a childlike plea: "I need to go . . . ."
"Okay." Shaltie nodded. Ray Shaltie wasn't a bad man. He'd worked the courts for two decades, and the experience had mellowed him—allowed him to see the human side of even the worst of men. And Bekker was the worst of men.
But Bekker was nevertheless human, Shaltie believed: _He that is without sin among you, let him cast the first stone._ . . . Bekker was a man gone wrong, but still a man. And in words that bubbled from his mouth in a whiny singsong, Bekker told Shaltie about his hemorrhoids. Jail food was bad for them, Bekker said. All cheese and bread and pasta. Not enough roughage. He had to go . . . .
He always used the bathroom at noon, all through the twenty-one days of the trial. Raymond Shaltie sympathized: he'd had them himself. Shaltie took Bekker by the arm and led him past the now empty jury box, Bekker shuffling, childlike, eyes unfocused. At the door, Shaltie turned him—docile, quiet, apparently gone to another world—and put on the handcuffs and then the leg chains. Another deputy watched the process, and when Bekker was locked up, drifted away, thinking of lunch.
"Gotta go," Bekker said. His eyes turned up to Ray Shaltie.
"You'll be okay, you'll be okay," Shaltie said. Shaltie's tie had soup stains on it, and flakes of dandruff spotted his shoulders: an oaf, Bekker thought. Shaltie led Bekker out of the courtroom, Bekker doing the jailhouse shuffle, his legs restricted to a thirty-inch stride. Behind the courtroom, a narrow hallway led to an internal stairway, and from there, to a holding cell. But to the left, through a service door, was a tiny employees-only men's room, with a sink, a urinal, a single stall.
Shaltie followed Bekker into the men's room. "Now, you're okay . . ." A warning in his voice. Ray Shaltie was too old to fight.
"Yes," Bekker said, his pale-blue eyes wandering in their sockets. Behind the wandering eyes, his mind was moving easily now, the adrenaline acting on his brain like a dose of the purest amphetamine. He turned, lifted his arms up and back, thrusting his wrists at Shaltie. Shaltie fitted the key, uncuffed the prisoner: Shaltie was breaking the rules, but a man can't wipe himself if he's wearing handcuffs. Besides, where would Bekker go, high up here in the government building, with the leg chains? He couldn't run. And his wildly bearded face was, for the moment at least, the most recognizable face in the Cities.
Bekker shuffled into the stall, shut the door, dropped his trousers, sat down. Eyes sharp now, focused. They used disposable safety razors in the jail, Bics. He'd broken the handle off one, leaving only the head and a stub, easy to hide during the shakedowns. When he'd had a chance, he'd burned the stub with a match, rounding the edges, to make it more comfortable to wear. This morning he'd taped it under his balls, fixed with the end of a Band-Aid. Now he peeled the razor off himself, pulled the remaining tape off the razor, and began hacking at his beard.
He'd grown the beard to cover his furrowed face. Bekker, once so beautiful, the possessor of a classic Nordic face, a pale, uninflected oval with rose lips, had been beaten into a grotesque gnome, torn to pieces and only poorly repaired. _Davenport. Get Davenport._ The fantasy seized him: opening Davenport, using the knife to peel the face, lifting the skin off inch by inch . . . .
He fought it: fantasies were for the lockup. He forced Davenport out of his mind and continued shaving, quickly, raggedly, the razor scraping over his dry skin. The pain prompted a groan. Outside the stall, Shaltie winced.
" 'Bout done in there?" Shaltie called. The bathroom smelled of ammonia, chlorine, urine, and wet mops.
"Yes, Ray." Bekker dropped the razor in his jacket pocket, then worked on the toilet-paper holder. Originally, it had been held in place with four screws. He'd removed and flushed two of them during the first three days of the trial, and had worked the other two loose. He'd actually had them out the day before, to make sure the holder would pull free. It had. Now he removed the screws one last time, dropped them in the toilet and eased the paper-holder free from the wall. When he grasped it by the roller, it fit his hand like a steel boxing glove.
"Okay now, Ray." Bekker stood, pulled his pants up, pulled off his jacket, dropped the coat over the iron fist, flushed the toilet. Took a breath. Put his head down, as though he were looking at his fly. Opened the door. Shuffled forward.
Shaltie was waiting with the cuffs: jowly, freckled, slow on the uptake. "Turn around . . . ."
Seeing Bekker's face, realizing: "Hey . . ."
Bekker was half-turned, wound up. He dropped the jacket, his right hand whipping like a lash, his mouth open, his white teeth flashing in the fluorescence. Shaltie lurched back, tried to cover with a hand. Too late, too late. The stainless-steel club hit him above the ear: Shaltie went down, cracking the back of his head on the porcelain sink as he fell.
And then Bekker was on him, lifting the steel fist, smashing it down, lifting it, feeling Shaltie's skull crack, the blood spatter.
_Hit hit hit hit_ . . .
The synapses of Bekker's brain lit with the static sparks. He fought it, fought for control, but it was hard, the smell of fresh blood in his nose. He stopped swinging, found his left hand on Shaltie's throat. Pulled the hand away, half stood, brain not quite right: He said aloud, shushing himself, "Shhh. Shhhhhh," finger to his lips.
He straightened. His blood was running like water now, like steam, filling him. Now what? Door. He hobbled to the door, flipped the catch. Locked. Good. He went back to Shaltie, who was supine on the tile floor, blowing blood bubbles through his torn nose. Bekker had watched the deputy handle his keys, and the keys had gone in Shaltie's right pocket . . . . He found them, popped the locks on the leg chains. Free. _Free._
Stop. He brought himself back, looked in the mirror. His face was a mess. He retrieved the razor from his jacket pocket, splashed water and liquid soap on his face and raked the razor across it. Listened to Shaltie, breathing, a gargling moan. Shaltie's head lay in a puddle of blood, and Bekker could smell it.
Bekker threw the razor in a trash basket, turned, stooped, caught Shaltie under the shoulders, dragged him to the toilet stall, sat him on the toilet and propped him against the wall. Shaltie made a snoring sound and more blood bubbled from his nose. Bekker ignored him. Not much time.
He stripped off his suit pants, put the Sox hat on his head, and used the pants to wipe up the blood on the floor. When he finished, he threw the pants, jacket, shirt and tie over Shaltie's body. Checked himself in the mirror: green tank top, red shorts, gym shoes, hat. A jogger. The face was bad, but nobody had seen him close up, without a beard, for weeks. A few of the cops would know him, a couple of lawyers. But with any luck, they wouldn't be looking at joggers.
_Davenport._ The thought stopped him. If Davenport was out there, had come to see the verdict, Bekker was a dead man.
No help for that. He threw off the thought, took a breath. Ready. He stepped inside the stall with Shaltie, locked it, dropped to his back, slid under the door, stood up again.
"Motherfucker." He said it out loud, had learned it in jail: the standard, all-purpose curse. He dropped back on the floor, slid halfway under the stall, groped for Shaltie's wallet. Found it, checked it. Twelve dollars. One credit card, a Visa. Not good. Money could be a problem . . . . He slipped the wallet into his underpants, went to the door, listened.
Could hear Shaltie breathing, bubbling. Bekker thought about going back into the stall, strangling him with his belt. All the humiliations of the past weeks, the torture when they took away his chemicals . . . Not enough time. Time was hurting him now. Had to move.
He left Shaltie living, turned the lock knob, peered into the hallway. The internal corridor was empty. Went to the next door—public hall. Half-dozen people, all down at the public end, near the elevators, talking. He wouldn't have to walk past them. The stairs were the other way: he could see the exit sign, just beyond the fire hose.
Another breath. And move. He stepped out into the hall, head down. A lunchtime bureaucrat-jogger on his way outside. He walked confidently down the hall to the stairs, away from the elevators. Waiting for a shout. For someone to point a finger. For running feet.
He was in the stairway. Nobody took the stairs, not from this high up . . . .
He ran down, counting the floors. As he passed six, a door slammed somewhere below and he heard somebody walking down ahead of him. He padded softly behind, heard another door open and shut, and stepped up the pace again. At the main level, he stopped and looked out. Dozens of people milled through the reception area. Okay. This was the second floor. He needed one more. He went down another level, and found an unmarked steel door. He pushed it open. He was outside, standing on the plaza. The summer sun was brilliant, the breeze smelled of popcorn and pigeons. A woman sitting on a bench, a kid next to her. She was cutting an apple with a penknife, her kid waiting for the apple.
Head down, Bekker jogged past her. Just another lunchtime fitness freak, weaving through the traffic, knees up, sweating in the sunshine.
Running like a maniac.
## CHAPTER
2
Lucas whipped down the asphalt backroads of Wisconsin, one hand on the wheel, one on the shifter, heel-and-toe on the corners, sunlight bouncing off the Porsche's dusty windshield. He slow-footed across the St. Croix bridge at Taylor's Falls into Minnesota, looking for cops, then dropped the hammer again, headed south into the sun and the Cities.
He caught Highway 36 west of Stillwater, the midday traffic sparse and torpid, pickups and station wagons clunking past the cow pastures, barns and cattail sloughs. Eight miles east of Interstate 694, he blew the doors off a red Taurus SHO. Clear road, except for the occasional crows picking at roadkill.
His eyes dropped to the speedometer. One hundred and seven.
_What the fuck are you doing?_
He wasn't quite sure. The day before, he'd rolled out of his lake cabin late in the afternoon and driven eighty miles north to Duluth. To buy books, he thought: there were no real bookstores in his corner of Wisconsin. He'd bought books, all right, but he'd wound up drinking beer in a place called the Wee Blue Inn at eight o'clock in the evening. He'd been wearing a dark-blue dress shirt under a silk jacket, khaki slacks, and brown loafers, no socks. A laid-off ore-boatman, drunk, had taken exception to the bare feet, and for one happy instant, before the barkeeps arrived, it had looked like the boatman would take a swing.
He needed a bar fight, Lucas thought. But he didn't need what would come afterward, the cops. He took his books back to the cabin, tried to fish the next day, then gave it up and headed back to the Cities, driving as fast as he knew how.
A few miles after blowing off the SHO, he passed the first of the exurban ramblers, outriders for the 'burbs. He groped in the glove box, found the radar detector, clipped it to the visor and plugged it into the cigarette-lighter socket as the Porsche screamed down the cracked pavement. He let his foot settle further; punched up the radio, Cities-97. Little Feat was playing hard hot boogie, "Shake Me Up," the perfect sound to accompany a gross violation of the speed limit.
The interstate overpass flicked past and the traffic got thicker. A hundred and eighteen. Hundred and nineteen. A stoplight he'd forgotten about, looming suddenly, with a blue sedan edging into a right-on-red turn. Lucas went left, right, left, heel-and-toe, blowing past the sedan; and past a station wagon, for a split second catching at the periphery of his vision the surprised and frightened face of a blond matron with a car full of blond kids.
The image fixed in his mind. Scared. He sighed and eased off the gas pedal, coasting. Dropped through a hundred, ninety, eighty. Across the northern suburbs of St. Paul, onto the exit to Highway 280. When he'd been a cop, he'd always been sneaking off to the lake. Now that he wasn't, now that he had time sitting on him like an endless pile of computer printout, he found the solace of the lake less compelling . . . .
The day was warm, sunshine dappling the roadway, playing games with cloud-shadows on the glass towers of Minneapolis to the west. And then the cop car.
He caught it in the rearview mirror, nosing out of Broadway. No siren. His eyes dropped to the speedometer again. Sixty. The limit was fifty-five, so sixty should be fine. Still, cops picked on Porsches. He eased off a bit more. The cop car closed until it was on his bumper, and in the rearview mirror he could see the cop talking into his microphone: running the Porsche's tags. Then the light bar came up and the cop tapped his siren.
Lucas groaned and rolled to the side, the cop fifteen feet off his bumper. He recognized him, a St. Paul cop, once worked with the Southwest Team. He used to come into the deli near Lucas' house. What was his name? Lucas dug through his memory. Kelly . . . Larsen? Larsen was out of the car, heavy face, sunglasses, empty-handed. No ticket, then. And he was jogging . . . .
Lucas shifted into neutral, pulled the brake, popped the door and swiveled in the seat, letting his feet fall on the shoulder of the road.
"Davenport, God damn it, I _thought_ this was your piece of shit," Larsen said, thumping the Porsche's roof. "Everybody's looking for your ass . . . ."
"What . . . ?"
"Fuckin' Bekker blew out of the government center. He's knocked down two people so far."
"What?" Lucas Davenport: deep summer tan, jagged white scar crossing his eyebrow, khaki short-sleeved shirt, jeans, gym shoes. A surge of adrenaline almost took his breath away.
"Two of your buddies are laying up at your place. They think he might be coming for you," Larsen said. He was a large man who kept hitching up his belt, and peering around, as though he might spot Bekker sneaking through a roadside ditch.
Lucas: "I better get my ass down there . . . ."
"Go." Larsen thumped the top of the car again.
Back on the highway, Lucas picked up the car phone and poked in the direct-dial number for the Minneapolis cops. He was vaguely pleased with himself: he didn't need the phone, rarely used it. He'd installed it the week after he'd bought the gold-and-steel Rolex that circled his wrist—two useless symbols of his freedom from the Minneapolis Police Department. Symbols that he was doing what every cop supposedly wanted to do, to go out on his own, to _make_ it. And now the business was snaking off in new directions, away from games, into computer simulations of police tactical problems. Davenport Games & Simulations. With the growing sales, he might have to rent an office.
The switchboard operator said, "Minneapolis."
"Gimme Harmon Anderson," Lucas said.
"Is that you, Lucas?" the operator asked. Melissa Yellow Bear.
"Yeah." He grinned. Somebody remembered.
"Harmon's been waiting. Are you at home?"
"No, I'm in my car."
"You heard what happened?" Yellow Bear was breathless.
"Yes."
"You take care, honey. I'll switch you over . . . ."
A moment later, Anderson came on, and said without preamble, "Del and Sloan are at your place. Sloan got the key from your neighbor, but they're wasting their time. He won't be coming after this long. It's been three hours."
"How about Del's place? He and Bekker are relatives of some kind."
"We've got a couple of guys there, too, but he's hiding somewhere. He won't be out, not now."
"How did he—"
"Go on home and Sloan can fill you in," Anderson said, interrupting. "I gotta go. This goddamn place is a madhouse."
And he was gone. _Police work to do, no time for civilians._ Lucas got off at University Avenue, took it to Vandalia, across I-94 and down Cretin, then over to the tree-shaded river road. Brooding. _No time for Davenport._
Feeling sorry for himself, knowing it.
Two blocks before he got to the house, he slowed, watching, then turned a block early. The neighborhood offered few places to hide, other than inside the houses. The yards were open, tree-filled, burning with color: crabapple blossoms and lines of tulips, banks of iris, pink peonies and brilliant yellow daffodils, and the odd patch of buttery dandelions that had somehow escaped the yard-service sprayers. The day was warm, and people were working in their yards or on their houses; a couple of kids in shorts shot baskets at a garage-mounted hoop. Bekker couldn't hide in the open yards, and breaking into a house would be tough. Too many people around. He turned a corner and idled down toward his house.
Lucas lived in what a real estate woman had once called a soft rambler: stone and clapboard, a fireplace, big trees, two-car garage. At the end of the asphalt drive, he slowed, punched the garage-door opener, and waited at the end of the driveway until the door was all the way up. A curtain moved in the front room.
When Lucas pulled into the garage, Sloan was waiting in the door between the house and the garage, hand in his jacket pocket. He was a thin man with high cheekbones and deep-set eyes. As Lucas got out of the car, Del drifted up behind Sloan, the butt of a compact 9mm pistol sticking out of his waistband. Del was older, with a face like sandpaper, a street burnout.
"What the hell happened?" Lucas asked as the garage door rolled down.
"An old-fuck deputy uncuffed him so he could take a shit," Sloan said. "Bekker'd been telling everybody that he had hemorrhoids and he always went to the can at the noon recess."
"Setting them up," Lucas said.
Del nodded. "Looks like it."
"Anyway, the jury went out and the deputy took him to the bathroom before hauling him down to the holding cell," Sloan continued. "Bekker unscrewed a steel toilet-paper holder from the wall of the stall. Came out of the stall and beat the shit out of the old guy."
"Dead?"
"Not yet, but he's leaking brains. He's probably paralyzed."
"I heard he hit two guys?"
"Yeah, but the other was later . . ." Del said, and explained. Witnesses waiting outside a courtroom had seen Bekker leave, without knowing until later who he was. Others saw him cross the government-center plaza, running past the lunchtime brown-baggers, through the rafts of pigeons, heading down the street in his shorts. "He went about ten blocks, to a warehouse by the tracks, picked up a piece of concrete-reinforcement rod, went in the warehouse and whacked a guy working at the dispatch desk. A clerk. Took his clothes and his wallet. That's where we lost him."
"The clerk?"
"He's fucked up."
"I'm surprised Bekker didn't kill him."
"I don't think he had time," Del said. "He's in a hurry, like he knows where he's going. That's why we came here. But it don't feel right anymore, the longer I think about it. You scare the shit out of him. I don't think he'd take you on."
"He's nuts," said Lucas. "Maybe he would."
"Whatever, you got a carry permit?" asked Sloan.
"No."
"We'll have to fix you up if we don't get him . . . ."
They didn't get him.
Lucas spent the next forty-eight hours checking old sources, but nobody seemed much inclined to talk to him, not even the cops. Too busy.
He brought a Colt Gold Cup .45 up from the basement gun safe, cleaned it, loaded it, kept it under the bed on a book. During the day, he carried it hidden in the Porsche. He enjoyed the weight of the gun in his hand and the headache-making smell of the gun-cleaning solvent. He spent an hour in a Wisconsin gravel pit, shooting two boxes of semi-wadcutters into man-sized silhouettes.
Then, two days after Bekker broke out of the courthouse, neighbors found the body of Katherine McCain. She'd been an antiques dealer and a friend of Bekker's wife, and she'd had the Bekkers to a party six or eight weeks before Bekker's wife had been murdered. Bekker knew the house and knew she lived alone. He'd been waiting when she came home, and killed her with a hammer. Before he left in her car, he'd used a knife to slash her eyes, so her ghost couldn't watch him from the other world.
And then he disappeared.
McCain's car was eventually found in an airport parking lot in Cleveland, Bekker long gone. On the day the car was found, Lucas put the .45 back in the gun safe. He never got the carry permit. Sloan forgot, and then after a while, it didn't seem important.
Lucas had temporarily gone off women, and found it hard to focus on the idea of a date. He tried fishing, played golf every day for a week. No good. His life, he thought with little amusement, was like his refrigerator—and his refrigerator contained a six-pack of light beer, three cans of diet caffeine-free Coke, and a slowly fossilizing jar of mustard.
At night, unable to sleep, he couldn't get Bekker out of his head. Couldn't forget the taste of the hunt, of closing in, of cornering him . . .
He missed it. He didn't miss the police department, with its meetings and its brutal politics. Just the hunt. And the pressure.
Sloan called twice from Minneapolis, said it looked like Bekker was gone. Del called once, said they'd have to get a beer sometime.
Lucas said _yeah._
And waited.
Bekker was a bad penny.
Bekker would turn up.
## CHAPTER
3
Louis Cortese was dying.
A brilliant floodlight lit his waxy face and the blood on his cheeks, and emphasized the yellow tint in his eyes. His lips were twisted, like those of an imp in a medieval painting.
Bekker watched. Touched a switch, heard the camera shutter fire. He could feel death swooping down on them, in the little room, in the lights, as Louis Cortese's life drained into a plastic jug.
Bekker's brain was a calculator, an empty vessel, a tangle of energy, a word processor, and an expert anatomist. But never more than one thing at a time.
Three months in the Hennepin County Jail had changed him forever. The jailers had taken away his chemicals, boiled his brain, and broken forever the thin electrochemical bonds that held his mind together.
In jail, lying in his cell in his rational-planner mode, he'd visualized his brain as an old-fashioned Lions Club gumball machine. When he put in a penny, he got back a gumball—but he never knew in advance what color he'd get.
The memory of Ray Shaltie, of the escape, was one color, a favorite flavor, rattling down the payoff chute of his psyche. When he got it, it was like a wide-screen movie with overpowering stereo sound, a movie that froze him in his tracks, wherever he was. He was _back there_ with Ray Shaltie, with the steel fist, smashing . . . .
Bekker, real time.
He sat in a chromed-steel chair and watched Cortese's death throes, his eyes moving between the monitor screens and the dying subject's face. A clear plastic tube was sewn into Cortese's neck, piping the blood from his carotid artery to an oversized water jug on the floor. The blood was purple, the color of cooked beets, and Bekker could smell it, his fine nostrils twitching with the scent. On the EKG, Cortese's heart rate soared. Bekker trembled. Cortese's consciousness was moving outward, expanding, joining with . . . what?
Well. Nothing, maybe.
Cortese's . . . essence . . . might be nothing more than a bubble reaching the top of a cosmic glass of soda water, expanding only to burst into oblivion. The pressure of the thought made Bekker's eyebrow jump uncontrollably, twitching, until he put a hand to his forehead to stop it.
There _had to be_ something beyond. That he himself might just blink out . . . No. The thought was insupportable.
Cortese convulsed, a full-body rictus throwing him against the nylon restraining straps, his head cranking forward, his eyes bulging. Air squeezed from his lungs, past the elaborate gag, a hoarse bubbling release. He was looking at nothing: nothing at all. He was beyond vision . . . .
The alarm tone sounded on the blood-pressure monitor, then on the EKG, twin tones merging into one. With his left hand still clapped to his forehead, restraining the unruly eyebrow, Bekker turned toward the monitors. Cortese's heart had stopped, blood pressure was plummeting toward zero. Bekker felt the large muscles of his own back and buttocks tighten with the anticipation.
He looked at the EEG, the brain-wave monitor. A jagged, jangled line just seconds before, it began flattening, flattening . . .
He felt Cortese go: could feel the _essence_ go. He couldn't measure it—not yet—but he could feel it. He bathed in the feeling, clutched at it; fired a half-dozen photos, the motor drives going _bzz-whit, bzz-whit_ behind his head. And finally the magic _something_ slipped away. Bekker jumped to his feet, frantic to hold on. He leaned over Cortese, his eyes four inches from the other's. There was something about death and the eyes . . . .
And then Cortese was gone, beyond Bekker's reach. His body, the shell of his personality, went slack beneath Bekker's hands.
The power of the moment spun Bekker around. Breathing hard, he stared at a reflection of himself in a polished stainless-steel cabinet. He saw himself there a dozen times a day, as he worked: the raw face, the sin face, he called it, the cornrows of reddened flesh where the gunsights had ripped through him. He said in a small, high voice: "Gone."
But not quite. Bekker felt the pressure on his back; his spine stiffened, and a finger of fear touched him. He turned, and the dead man's eyes caught him and held him. They were open, of course. Bekker had carefully trimmed away the eyelids to ensure they would remain that way.
"Don't," he said sharply. Cortese was mute, but the eyes were watching.
"Don't," Bekker said again, louder, his voice cracking. Cortese was watching him.
Bekker snatched a scalpel from a stainless-steel tray, stepped to the head of the table, leaned over the body and slashed at the eyes. He was expert: it only took a second. He carved the eyes like boiled eggs, and the vitreous aqua leaked down Cortese's dead cheeks like jellied tears.
"Good-bye," Bekker said dreamily. The ruined eyes were no longer threatening. A gumball dropped, and Bekker went away . . . .
Thick stopped at the curb, rocking on his heels, waiting patiently for the light. Thin snapped a cigarette into the street, where it exploded in a shower of sparks. The cars went by in a torrent, battered Toyotas and clunking Fords, fender-bent Dodges, pickups and vans blocking the view ahead, trucks covered with graffiti, buses stinking up the streets with noxious diesel fumes, all rolling past like iron salmon headed upstream to spawn. Through all of it, the taxis jockeyed for position, signaling their moves with quick taps on their horns, an amber warp to the woof of the street. New York was noise: an underground rumble of trains and steam pipes, a street-level clash of gears and motors and bad mufflers, a million people talking at once, uncounted air conditioners buzzing above it all.
All of it congealed in the heat.
"Too fuckin' hot," Thick said. And it was; he could feel it on his neck, in his armpits, on the soles of his feet. He glanced at Thin, who'd stopped at the curb beside him. Thin nodded but didn't answer. They were both wearing long-sleeved shirts with the sleeves rolled down to their wrists. Thin was a problem, and Thick didn't quite know what to do about it. Hadn't really known, he thought wryly, for almost forty years now . . . .
The walk sign flashed on and he and Thin crossed the street. A traffic-light pole, splattered with pigeon shit and encrusted with the grime of decades, sat on the corner. At the bottom, and up as high as a hand could reach, it was covered with fading posters. Above that, two street signs were mounted at right angles to each other, a bus-stop sign faced the street, and a temporary traffic-diversion sign pointed an arrow to the left. Above all that, a spar went out to the traffic signal, and another supported a streetlight.
_Oughta put one in a fuckin' museum someplace, just like that. Our own fuckin' totem poles_ . . .
"Dollar . . ." The woman on the sidewalk reached up at him, holding a dirty hand-lettered card: "Help me feed my children." Thick walked past, thinking that it was impossible that the woman had children. In her forties, perhaps, she was withered as a week-old carrot, her emaciated legs sprawled beneath her, her bare feet covered with open sores. Her eyes had a foggy-white glaze, not cataracts, but something else. She had no teeth at all, only dimples in gray gums, like the vacant spots left by corn kernels popped from a cob.
"I read this book about Shanghai once, the way it was before World War Two," Thick said as they passed on. Thin looked straight ahead, not responding. "The thing was, begging was a profession, you know? But an ordinary guy couldn't get any alms. You needed to be special. So they'd take kids and burn their eyes out or smash their arms and legs with hammers. They had to make them pitiful enough to get money in a whole city full of beggars . . . ."
Thin looked up at him, still saying nothing.
"So we're getting there, too," Thick said, looking back at the woman on the street. "Who's gonna give money to your average panhandler when you walk by something like that every day?" He half turned to look back at the woman.
"Dollar," the woman wailed, "Dollar . . ."
Thick was worried. Thin was talking about running out. He glanced at his partner. Thin's eyes were angry, fixed straight ahead. Thinking . . .
Thick was carrying a large, flat, cardboard box. It wasn't particularly heavy, but the shape was awkward, and he slowed to hitch it up under his arm.
"I wouldn't mind . . ." Thick started, then let it go. He reached up to scratch his face, but he was wearing thin, flesh-colored surgeon's gloves, and he couldn't effectively scratch. They moved along, quickly, to an apartment building across the street from the steak house. Thick had the key in his free hand and opened the door.
Thin said, "I can't do it."
"We gotta. Jesus Christ, if we don't we're fuckin' dead, all of us . . . ."
"Listen . . ."
"Off the street, off the street . . ."
Inside the door, the hall and landing were dimly lit by a yellow sixty-watt bulb. The stairs were immediately to the right, and Thick started up. Thin, undecided, looked back out at the street, then, reluctantly, because Thick was already moving, followed. At the top of the stairs, they stopped in the hallway for a moment and listened, then went to the front apartment and opened the door with a key. The only light in the apartment came through the yellowed shades on the front windows, from the street. The place smelled of dead air, old coffee grounds, and dry plants. The owners had been in Rome for a week, to see the Pope. They'd go to the Holy Land afterward. The Holy Land in July. They'd burn their brains out, if they had any, which they probably didn't, if they were going to the Holy Land in July.
Thin shut the door behind them and said, "Listen . . ."
"If you weren't going to do it, why'd you come this far?"
"Because you got us into it. I don't want you to get fucked up."
"Jesus . . ." Thick shook his head and stepped carefully through the dark room to the windows and lifted a shade. "Get the rifle."
"I'm not . . ."
"All right, I'll do it. Jesus, if that's the way you feel about it, go. Get the fuck out," Thick said, anger riding his voice. He was older than Thin by twenty-three years and two days, his face stamped with the cuts and gullies of a life on the street. He picked up the box he'd carried in. "Go."
Thin hesitated, watching. The box was five feet long by three wide, but only eight inches deep. It might have held a mirror, or even a painting, but it didn't—it held a Colt AR-15 with a flash suppressor, a twenty-shot magazine, a two-power light-gathering scope, and a laser sight. The weapon, manufactured as a semiautomatic, had been converted to selectible fire, semiauto or full auto, by a machinist in Providence.
Thick had spent an afternoon in the Adirondacks shooting plastic milk jugs from a perch high on the bank of a gully. The gallon-sized jugs closely simulated the kill zone of a man's chest from any angle. Thick used hand-loaded cartridges, and he was a very good shot. When hit by one of Thick's hot loads, the milk jugs literally exploded.
Thick used a penknife to cut the twine that held the box shut, stripped off a couple of pieces of tape, opened it, and took the weapon out of the sponge-rubber packing. New scope mounts weren't as delicate as those he'd grown up with, but there was no point in taking chances. He hadn't. A fully loaded magazine was packed with the weapon. Each cartridge had been polished with a chamois to eliminate fingerprints. Thick slapped the magazine home with his rubber-gloved hands.
"Get the couch," Thick said. "Hurry it up."
"No: he's a cop. If he wasn't a cop . . ."
"Bullshit." Thick went to the windows, looked out on the empty street, then unlocked one of them and carefully raised it until it was fully open. Then he turned, glanced at Thin, and picked up the rifle.
"You never had this problem before . . . ."
"The guy hasn't done anything. The others were scumbags . . . . This is a cop . . . ."
"He's a goddamn computer asshole cockroach and he's gonna put good guys in jail for doing what had to be done. And you know what happens if we get sent up? We're fuckin' dead, that's what. I personally doubt that I'd last a fuckin' week; if they come for me, I'm stickin' my goddamn pistol in my mouth, because I ain't goin' . . . ."
"Jesus . . ."
Thick, standing well back from the window, looked at the restaurant across the street through the low-powered scope. A Visa emblem was stuck to the window on the door, under the script of the restaurant's name and logo. Looking at the logo, the theme song from an old television show trickled through his head: _"Have gun, will travel" is the card of a man_ . . .
He picked up the Visa sign in the scope, touched the laser switch with his thumb. A red dot bloomed on the sign. Thick had a head the size of a gasoline can, with small ears that in the semidark looked like dried apricots. "He's worse than the shooflies."
"He . . ." Thin's eyes went to the street, and Thick followed them. The restaurant door was opening.
"Wrong guy," Thin blurted.
"I know . . . ."
A man in a white tennis shirt and white shoes stood there, probing his gums with a plastic toothpick. The toothpicks were shaped like swords, Thin knew. They'd made a recon trip to the steak house the night before, to figure times and placements. The target always came in for the Friday special, New York strip with sour-cream baked potato and choice of draft beer. The man in the tennis shirt ambled down the street.
"Fuckin' faggot," said Thick. He flicked the switch on the laser sight and the red dot bloomed on the Visa sign.
Bekker sighed.
All done.
He turned away from Cortese's body, his mind like a coil of concertina wire, tense, sharp, dangerous. He touched his shirt pocket: the pocket was empty. He stepped out of his room, with a touch of anxiety, and went to the old dresser where he kept his clothes. A half-handful of pills were scattered across the top of it, and he relaxed. Enough. He picked up several, developing a combo rush as he went, popped them into his mouth, savored the acrid bite, and swallowed. So good; but so few. He looked at the top of the dresser, at the pills there. Enough for another day, no more. He'd have to think about it—but later.
He went back into the workroom, killed the monitors, their green screens blanking out. Nothing to see anyway, just horizontal lines. Bekker ignored the body. Cortese was simply garbage, a matter of disposal.
But before the death . . . A new gumball dropped, and Bekker froze beside the worktable, his mind sliding away.
Louis Cortese had been dark-haired, seventy-one and one-half inches tall, one hundred and eighty-six pounds, and thirty-seven years old—all of it carefully recorded in Bekker's notebooks. He'd been a graduate in electrical engineering from Purdue University. Before Bekker'd cut off his eyelids, when Cortese had still been trying to ingratiate himself, still fending off the idea that he was about to die, he'd told Bekker that he was a Pisces. Bekker had only a vague idea what that meant, and he wasn't interested.
Cortese's body lay on a stainless-steel countertop, which had cost six hundred and fifty dollars at a restaurant-supply shop in Queens. The countertop, in turn, was fixed to an old wooden library table; Bekker'd had to cut down the legs to get the proper working height. Overhead, a rank of three shop lamps threw a flat, cold light on the table.
Because his research subjects would be alive, Bekker had fixed restraining rings to the table. A brown nylon strap was clipped to a ring just below Cortese's right armpit, and ran diagonally from the armpit across the chest between the nipple and the shoulder, to another ring behind the neck, then from behind the neck, back across the opposite side of the chest to another ring below the left armpit; it held Cortese like a full nelson. Additional straps crossed the body at the waist and knees and bound the wrists and ankles.
One of the hands was taped as well as bound: Bekker monitored blood pressure through a catheter placed in the radial artery, and the wrist had to be totally immobilized. Cortese's jaws were spread wide, held open by a hard-rubber cone: the subject could breathe through the nose, but not through the mouth. His screams, when he tried to scream, sounded like a species of humming, though not quite humming.
Mostly, he'd been as silent as a book.
At the head of the table, Bekker had stacked his monitoring equipment in what a discount stereo store had called a home entertainment center. The arrangement was pleasingly professional. The monitors measured body temperature, blood pressure, heartbeat, and brain-wave activity. He also had a neuro-intracranial pressure monitor, but hadn't used it.
The room around the equipment was also carefully finished: he'd worked on it for a week before he was satisfied. Scrubbed it with disinfectant. Installed an acoustic-tile ceiling and Formica wall panels in a smooth oyster-white finish. Put down the royal-blue carpet. Brought in the equipment. The monitors had been the hard part. He'd finally gotten them from Whitechurch, a dealer at Bellevue. For two thousand in cash, Whitechurch had taken them out of a repair shop, first making sure they'd been fixed . . . .
Sigh.
One of the monitors was telling him something.
What was it? Hard to concentrate . . .
Body temperature, eighty-four degrees.
_Eighty-four?_
That was too low. He glanced at the clock. 9:07 . . .
He'd been gone again.
Bekker rubbed the back of his neck, disturbed. He would go away, sometimes for an hour. It never seemed to happen at critical times, but still: he should have recognized it, the sigh when he came back. When he went away, he always came back with a sigh . . . .
He stepped to the tape recorders, looked at the counters. They were slightly out of sync, one of them at 504, the other at 509. He rewound them to 200 and listened to the first.
" . . . _direct stimulus brings only a slight reaction, no more than one millimeter_ . . ."
His own voice, hoarse with excitement. He turned off the first recorder, turned on the second. " . . . _no more than one-millimeter reflex in the iris followed by immediate release of_ . . ."
He turned off the second one. The recorders were working fine. Identical Sonys, with battery backup in case of power failure, they were better than the ones he'd used at the University of Minnesota.
Bekker sighed, caught himself, looked quickly at the clock, afraid that he'd been away again. No. 9:09. He had to clean up, had to get rid of Cortese's body, had to process the Polaroid color-slide film in the cameras. And he had some ideas about the taking of the specimens, and those ideas should be noted. Many things to do. But he couldn't, not at just this moment. The PCP hadn't arrived, and he felt . . . serene. The session had been a good one.
Sigh.
He glanced at the clock, felt a tiny thrill of fear. Nine twenty-five. He'd been gone again, frozen in one place; his knees ached from the unmoving stance. It was happening too much. He needed more medication. Street cocaine was good, but not precise enough . . . .
Then: _Dink._
Bekker turned his head. The intrusive sound came from a corner of his basement apartment. Almost a bell, but not quite. Instead of ringing, it simply struck once each time the old woman pushed the button.
_Dink._
Bekker frowned, walked to the intercom, cleared his throat, and pressed the talk button. "Mrs. Lacey?"
"My hands hurt." Her voice was shrill and ragged. Old. She was eighty-three, hard of hearing, nearly blind in one eye. Her arthritis was bad and growing worse. "My hands hurt so much," she complained.
"I'll bring a pill . . . in a few minutes," Bekker said. "But there are only three left. I'll have to go out again tomorrow . . . ."
"How much?" she asked.
"Three hundred dollars . . ."
"My golly . . ." She seemed taken aback.
"It's very difficult to find these days, Mrs. Lacey," Bekker said. And it had been for decades. She knew that. Morphine had never been street-legal in her lifetime. Neither had her marijuana.
A few days after he'd taken the job as a live-in helper—the old woman's word, she didn't need bathroom assistance—he'd shown her a _Wall Street Journal_ story about bank failures. She'd read it, nearly whimpering. She had her Social Security, she had her savings, some $370,000, and she had her building. If any of them broke down . . .
Edith Lacey had watched the old street women as they went by, pushing their shopping carts along the broken pavement, guarding their bundles of rags. She knew them, she said, although Bekker didn't believe her. She'd look out and make up stories about them. "Now that one, _she_ once owned a grocery on Greenwich . . ."
Bekker suggested that she spread her cash among three or four unrelated banks, so more would be insured by the Federal Deposit Insurance Corporation.
"Uncertain times," he told her in his careful voice.
She'd talked to her only ambulatory friend about it. Bridget Land, who didn't like Bekker, had thought that spreading the money among banks would be a good idea. And she'd volunteered to go with them: "To make sure everything is on the up-and-up," she'd said, her eyes moving almost involuntarily toward Bekker. "At the banks, I mean."
They'd moved the money in a single day, the two old women nervously guarding the cashier's checks like mother hens. Edith Lacey carrying one inside her blouse, Bridget Land the other in a buttoned pocket, _just in case._ They'd focused so closely on the checks that neither had paid much attention to Bekker as he reviewed Edith's applications for new accounts. Bekker had simply checked the "yes" box that asked if the applicants wanted automatic-teller cards. He picked up the mail each afternoon; a week after they'd moved the money to the banks, he'd intercepted the automatic-teller codes, and a week after that, the cards themselves. The cards were each good for five hundred dollars a day. During the first month, Bekker worked the accounts almost daily, until he had twenty thousand in cash.
"Get fruit," the old woman ordered.
"I'll stop at MacGuire's," he said on the intercom.
"Apricots."
"Okay." He started to turn away.
"Be sure to get apricots . . . ."
"Yes," he snapped.
"You didn't get them last time . . . ."
He was seized by a sudden urge to go up and choke her: not the urge that took him to his subjects, but an almost human desire to choke the shit out of a common nag. "I'm sorry," he said, abjectly, hiding the sudden fury. "And I'll _try_ to get your pills."
That would shut her up . . . .
Bekker turned away from the intercom and, through the dark living quarters, saw Cortese's body in the bright light flowing from the operating room. _Might as well do it now._
From the kitchen, he brought a long roll of black polyethylene, sold as painter's dropcloth. He unrolled it beside the dissection table, used a scalpel to cut it to the right length, then unfolded it. Unstrapped the body. Pulled the catheter from the wrist, pulled the temperature probe. The temperature was down to seventy-nine. Cooling quickly.
Bodies are hard to manipulate, and Bekker, with much experience, didn't even try. He simply walked to the far side of the table and pushed. The body rolled out of the tray and fell on the plastic sheet with a meaty thwack. He walked back around the table, wrapped it, folded the extra length, tied it with clothesline. He took two extra loops at the waist, to use as a handle. He hauled the body through the living quarters and up the steps to the building's reinforced back door, struggling with it. Even when you didn't care if they were damaged, bodies were difficult. And Cortese had been hefty. He should go after smaller people . . . .
The back door of the Lacey building was hidden from the street by a lean-to structure, designed as a car shelter. He popped open the door, chain still on, and checked the lean-to. In the past, bums had sheltered there. Nothing but the Volkswagen, undisturbed. He dragged the body outside, and, with some difficulty, stuffed it into the passenger seat. When it was in, he stepped to the edge of the lean-to and peeked toward the street. Nobody. He went back inside, closed the door, and hurried down the stairs.
Bekker showered, shaved carefully, dressed, and put on his makeup. The process was intricate: The heavy base makeup covered his ruined face, but had to be carefully shaded into the clear skin at his temples without obvious lines. He took half an hour, working at it. He'd just finished when Mrs. Lacey rang again.
"What?" _Old hag_ . . .
"My hands," she whined.
"I'm coming now," he said. Maybe he should kill her, he thought. He allowed himself to feel the pleasure of the idea. But then he'd have to explain her absence to Bridget Land. Though he _could_ eliminate Land . . . But that led into a maze of unresolvable questions and dangers: Did Land have other friends, and did they know she came to see Edith Lacey? If Land disappeared, would others come looking for her?
Killing her would be dangerous . . . . No, he would kill neither of them. Not yet. Lacey was the perfect front and Land was, so far, only a modest inconvenience. Bekker, thinking about them, got a bottle of pills from his bureau, shook one into the palm of his hand, went to the bottom of the stairs, flicked a light switch, and went up.
The stairs emerged into the back part of the first floor, then curled and went up to the second and third floors. The first floor had once been a plumbing-parts supply business, but had been vacant for years. During the day, a murky green light filtered through from the street. At night, the grille-covered windows were simply dark panels on either side of the street door.
The old woman huddled on the second floor, where she'd lived with her two cats since her husband's death. The second floor reeked of the three of them: cooked carrots, dope, and cat piss. Bekker hated the cats. They knew what he was and watched him from shelves, their eyes glittering in the gloom, as the old woman huddled in front of the television, wrapped in her tie-dyed shawl.
The third floor had once been part of the living quarters, when Mrs. Lacey's husband was living, but now, like the first, was vacant.
Bekker climbed to the second floor, the smell of carrots and marijuana closing around him. "Mrs. Lacey?"
"In here." She was a small woman, with thick glasses that enlarged her rheumy blue eyes. Her hair, wiry and gray, clung close to her head. She had a small button nose and tiny round lips. She was wrapped in a housecoat. She had four of them, quilted, in different pastel colors. She was waiting in the big chair in the living room, facing the television. Bekker went to the kitchen, ran a glass of water and carried the pill out to her. A cat ran from under her chair and hid in the next room, looking back at Bekker with cruel eyes.
"This'll help. I'll get more tomorrow."
"Thank you." She took the pill and drank greedily from the glass.
"You have your pipe and lighter?"
"Yes."
"You have enough of your tea?"
"Yes, thank you kindly." She cackled. She'd washed out of the bohemian life of the forties, but she still had her tea.
"I'm going out for a while," he said.
"Be careful, it's dangerous this late . . . ."
Bekker left her in her chair and went back down the stairs and carefully checked the lean-to again. Nobody.
The Lacey building fronted on Greene Street. The buildings on either side ran all the way back to Mercer, but the Lacey building filled only half the lot. The back lot, overgrown weeds and volunteer sumac, was closed off with a ten-foot chain-link fence. Before Bekker had arrived, vandals and bums had been over and through it and had broken the lock on the gate. After Bekker had bought the Volkswagen, he'd had the fence fixed and a long twisty strand of razor wire laid along the top.
Now he backed the Volkswagen out of the lean-to, wheeled it to the fence, hopped out, opened the gate, drove through, stopped once more, and locked the gate again.
_New York,_ he thought.
_Bagels and lox/Razor wire and locks._
Bekker giggled.
"Door," said Thick. He was standing by the window, the M-15 at his shoulder.
On the street below, an old-fashioned Volkswagen, a Bug, zipped past. Thick, looking through the scope, ignored it. A man had stepped out on the street and paused. He had light hair, slightly mussed, and gold-rimmed glasses. Narrow shoulders. He was smiling, his lips moving, talking to himself. He was wearing a blue short-sleeved shirt, and jeans that were too long for his legs. He used his index fingers to push his glasses up on his nose.
"Yes," Thick grunted, his finger tightening on the trigger.
"No . . ." said Thin, taking two steps toward the window.
But a red dot bloomed on the target's chest. He may have had an instant to think about it; again, maybe not. The blast of the gun was deafening, the muzzle flash brighter than Thin had expected. The target seemed to jump back, and then began a herky-jerky dance. Thin had once seen a film showing Hitler dancing a jig after the fall of France. The man on the street looked like that for just a second or two: as though he were dancing a jig. The thunder rolled on, six shots, eight, twelve, quick, evenly spaced, the lightning flickering off their faces.
A little more than halfway through the magazine, Thick flicked the selector switch and unloaded the remaining cartridges in a single burst. The target was now flat on the sidewalk, and the burst of bullets splattered about his head like copper-jacketed raindrops.
Thin stood by the window, unspeaking.
"Go," said Thick. He dropped the rifle on the floor. "Hands."
With their gloved hands pressed to their faces, they walked down the hall to the back of the building, ran down a flight of stairs, along another hallway, then out a side door into an alley. The alley led away from the shooting.
"Don't run," said Thick as they emerged onto the street.
"Watch it," said Thin.
A Volkswagen lurched past, a Bug, catching them in its lights, their pale faces like street lamps in the night. It was the same car that had driven past the restaurant just before the computer fag came out on the sidewalk . . . .
With the body beside him, Bekker was tense, cranked, watching for cop cars, watching everything that went by. He had a small pistol by his side, a double-barreled derringer .38 Special, but if he had to use it, he'd probably be finished.
But so far, so good.
SoHo streets were quiet at night. Once out of the neighborhood, things would get more complicated. He didn't want anything high beside him, a van or a truck. He didn't want a driver looking down into the Volkswagen, even though he probably wouldn't see much. The body, wrapped in dark plastic, looked more like a butterfly's chrysalis than anything, a cocoon. What you might expect from a Bug.
Bekker almost laughed. Not quite; he was too crazy to have a genuine sense of humor. Instead he said, "Motherfucker."
He needed a wall, or an unguarded building with a niche in the wall. Some place where nobody would look out and see him unloading the body. He hadn't thought much about disposal: he'd have to think more. He'd need a random dispersal pattern, nothing they could use to focus on his particular block. He'd have to decide the optimum distance—far enough not to point at SoHo, but not so far that the drive itself became risky.
He drove past the Manhattan Caballero, a Village steak house, a couple of bright beer signs in the small barred windows. The door opened as he went by and he saw a slender man come out, caught just for a moment by the light inside the doorway; and behind him, a cigarette machine.
The gunshots sounded like popcorn. Or like a woman ripping a piece of dress material. Bekker looked in the mirror, saw the lightning. Bekker had been in Vietnam; he'd heard this noise from a distance, this snickering popcorn thunder. He'd seen this flickering light. The man he'd seen in the doorway was flopping on the sidewalk as the bullets tore through him.
"Motherfucker . . ." Teeth bared, mouth wide, Bekker screamed the word: he was innocent, he had nothing to do with it, and he could get caught, right here. Half panicked, afraid that neighbors would take the number of every car they saw, Bekker floored the accelerator and raced to the end of the long block. The gunfire lasted for only two or three seconds. It took another five before he could turn left, out of sight, onto a one-way street. The adrenaline surged through him, the PCP panic. And up ahead, yellow lights flashed in the street.
_What?_
The panic jumped him. He jammed on the brake, forgetting the clutch, and the Volkswagen stalled. The body crinkled its plastic coat as it swayed in the seat toward him. He pushed it back with one hand, fighting the fist in his throat, trying to breathe, trying to get some air, and stabbed at the gas pedal. Finally realizing what had happened, he dropped the clutch and turned the key again, got started, shifting into second.
He jerked the car to the left, still dazzled, before he realized that the yellow lights were road-construction warnings. No reason to turn—but he already had, and he sped on. Near the end of the block, two figures stepped out of an alley. His headlights swept them, and he saw their hands come up. They were hiding their faces, but before they'd covered them, they'd been as clear as the face of the moon.
Bekker swerved, kept going.
Had they seen his plates? No way to tell. He peered into the rearview mirror, but they were already lost in the dark. He was okay. He tried to choke down the fear. The back plates were old and dirty.
But the gunfire.
Had to think. Jesus, he needed help. He felt for the matchbox. No, that wouldn't be right. He needed speed. Uppers, to help him think.
Sirens.
Somewhere behind him. He wasn't sure quite where he was anymore, took a left, moving away, coming up to a major intersection. He looked up at the street signs. Broadway. What was the other? He rolled forward a few feet. Bleecker. Okay. Good. Straight ahead, along Bleecker. Had to get the body out. A darker block, a deep-red building with niches, but no place to pull over. Another fifty feet . . . there.
He pulled to the curb, hopped out, and looked around. Nobody. He could hear somebody talking, loud, but it sounded like a drunk. He hurried around the car, shifted the body out and dropped it in a doorway. Looked up: the ceiling in the deep doorway was decorated with intricate designs in white terra-cotta; the designs caught his mind, dragged it into the maze of curves . . . .
Another siren brought him back. It was somewhere down Bleecker, but he couldn't see the lights. He hurried back to the car, sweating, climbed inside, and looked back through the open door at the mortal remains of Louis Cortese. From any more than a few feet, the body looked like a bum sleeping on the sidewalk. And there were hundreds of bums in the area.
He risked a last look at the terra-cotta, felt the pull, then tore his eyes away and slammed the door. Hunched over the steering wheel, he headed for home.
Thick picked up the pay phone and dialed the number scrawled on a scrap of paper. He let the phone ring twice, hung up, waited a few seconds, dialed again, let it ring twice more, hung up again.
Thin was waiting in the car, didn't speak.
"It'll be okay," Thick said.
After a very long time, Thin said, "No, it won't."
"It's fine," the big man said. "You did good."
When Bekker got to the Lacey building, he parked the car, went down into the basement, stripped off his clothes, scrubbed his face, changed into a sweat suit. And thought about the killing he'd seen. New York was a dangerous place—someone really ought to do something about it . . . . There was some cleanup to do in the operating theater. He worked at it for ten minutes, with a sponge and paper towels and a can of universal cleaner. When he was done, he wrapped all the paper and put it in the garbage. He remembered the blood just as he was about to turn out the lights. He picked up the bottle and tipped it into a drain, the blood as purple and thick as antifreeze.
Again he reached for the lights, and saw the four small nubbins of skin sitting on top of an anesthetic tank. Of course, he'd put them there, just a convenient place at the time.
He picked them up. Shriveled, with the long shiny lashes, they looked like a new species of arachnid, a new one-sided spider. They were, of course, something much more mundane: Cortese's eyelids. He peered at them in the palm of his hand. He'd never seen them like this, so separate, so disembodied.
Ha. Another one. Another joke. He looked in the stainless-steel cabinet, laughed and held his belly, and pointed a finger at himself. _Disembodied_ . . .
He went back to them, the eyelids. Fascinating.
## CHAPTER
4
Lucas was lying on the roof of his house, the shingles warm against his shoulder blades, eyes closed, not quite snoozing. He'd put down one full flat of green fiberglass shingles and didn't feel like starting another. A breeze ruffled the fine black hair on his forearms; the humid air was pregnant with an afternoon storm and pink-and-gray thunderheads were popping up to the west.
With his eyes closed, Lucas could hear the after-work joggers padding along the sidewalk across the street, the rattle of roller blades, radios from passing cars. If he opened his eyes and looked straight up, he might see an eagle soaring on the thermals above the river bluffs. If he looked down, the Mississippi was there, across the street and below the bluff, like a fat brown snake curling in the sunshine. A catsup-colored buoy bobbed in the muddy water, directing boat traffic into the Ford lock.
It all felt fine, like it could go on forever, up on the roof.
When the taxi pulled into the driveway, he thought about it instead of looking to see who it was. Nobody he knew was likely to come calling unexpectedly. His life had come to that: no surprises.
The car door slammed, and her high heels rapped down the sidewalk.
Lily.
Her name popped into his head.
Something about the way she walked. Like a cop, maybe, or maybe just a New Yorker. Somebody who knew about dog shit and cracked sidewalks, who watched where she put her feet. He lay unmoving, with his eyes closed.
"What are you doing up there?" Her voice was exactly as he remembered, deep for a woman, with a carefully suppressed touch of Brooklyn.
"Maintaining my property." A smile crept across his face.
"You could have fooled me," she said. "You look like you're asleep."
"Resting between bouts of vigorous activity," he said. He sat up, opened his eyes and looked down at her. She'd lost weight, he thought. Her face was narrower, with more bone. And she'd cut her hair: it had been full, to the shoulders. Now it was short, not punk, but asymmetrical, with the hair above her ears cut almost to the skin. Strangely sexy.
Her hair had changed, but her smile had not: her teeth were white as pearls against her olive skin. "You're absolutely gorgeous," he said.
"Don't start, Lucas, I'm already up to my knees in bullshit," she answered. But she smiled, and one of her upper incisors caught on her lower lip. His heart jumped. "This is a business trip."
"Mmmm." Bekker. The papers were full of it. Six already dead. Bodies without eyelids. Cut up, in various ways—not mutilated. Bekker did very professional work, as befitted a certified pathologist. And he wrote papers on the killings: strange, contorted, quasiscientific ramblings about the dying subjects and their predeath experiences, which he sent off to scientific journals. "Are you running the case?"
"No, but I'm . . . involved," she said. She was peering up at him with the comic helplessness with which people on the ground regard people on roofs. "I'm getting a crick in my neck. Come down."
"Who'll maintain my property?" he teased.
"Fuck your property," she said.
He took his time coming down the ladder, aware of the special care: _Five years ago, I'd of run down . . . hell, three years ago . . . getting older. Forty-five coming up. Fifty still below the horizon, but you could see the shadow of it . . ._
He'd been stretching, doing roadwork, hitting a heavy bag until he hurt. He worked on the Nautilus machines three nights a week at the Athletic Club, and tried to swim on the nights he didn't do Nautilus. Forty-four, coming onto forty-five. Hair shot through with gray, and the vertical lines between his eyes weren't gone in the mornings.
He could see the two extra years in Lily as well. She looked tougher, as though she'd been through hard weather. And she looked hurt, her eyes wary.
"Let's go inside," he said as he bent to let her kiss him on the cheek. He didn't have to bend very far; she was nearly as tall as he was. Chanel No. 5, like a whiff of distant farm flowers. He caught her by the arm. "Jesus, you look good. Smell good. Why don't you call?"
"Why don't you?"
"Yeah, yeah . . ." He led the way through the front door to the kitchen. The kitchen had been scorched in a gunfight and fire two years past, a case he'd worked with Lily. He'd repainted and put in a new floor.
"You've lost some weight," he said as they went, groping for something personal.
"Twelve pounds, as of this morning," she said. She dropped her purse on the breakfast bar, looked around, said, "Looks nice," pulled out a stool and sat down. "I'm starving to death."
"I've got two cold beers," Lucas said. He stuck his head in the refrigerator. "And I'm willing to split a deli roast beef sandwich, heavy on the salad, no mayonnaise."
"Just a minute," Lily said, waving him off. He shut the refrigerator door and leaned against it as she took a small brown spiral notebook from her purse. She did a series of quick calculations, her lips moving. "Airline food can't be much," she said, more to herself than Lucas.
"Not much," he agreed.
"Is it light beer?"
"No . . . but hell, it's a celebration."
"Right." She was very serious, noting the calories in the brown notebook. Lucas tried not to laugh.
"You're trying not to laugh," she said, looking up suddenly, catching him at it. She was wearing gold hoop earrings, and when she tipped her head to the side, the gold stroked her olive skin with a butterfly's touch.
"And succeeding," he said. He tried to grin, but his breathing had gone wrong; the dangling earring was hypnotic, like something out of a magician's show.
"Christ, I hate people with fast metabolisms," she said. She went back to the notebook, unaware of his breathing problems. _Maybe._
"That's all bullshit, the fast-metabolism excuse," Lucas said. "I read it in the _Times._ "
"Another sign of decline, the _Times_ printing obvious bullshit," Lily said. She stuffed the notebook back in her purse, put the purse aside and crossed her legs, clasping her hands on her knees. "Okay, a beer and half a sandwich."
They ate at the breakfast bar, facing each other, making small talk, checking each other. Lucas was off the police force and missed the action. Lily had moved up, off the street, and was doing political work with a deputy commissioner. Lily asked, "How's Jennifer? And Sarah?"
Lucas shook his head, finishing the sandwich. "Jen and I—we're all done. We tried, and it didn't work. Too much bad history. We're still friends. She's seeing a guy from the station. They'll probably get married."
"He's okay?"
"Yeah, I guess," Lucas said.
But he was unconsciously shaking his head as he said it.
Lily considered the tone: "So you think he's an asshole?"
"Hell . . . No. Not really." Lucas, finished with his half of the sandwich, stepped over to the sink, squirted Ivory Liquid into the palm of one hand, turned on the water and washed off the traces of the sandwich's olive oil. His hands were large and square, boxer's hands. "And he likes Sarah and he's got a kid of his own, about seven months older than Sarah. They get along . . . ."
"Like a family . . ." Lily said. Lucas turned away and shook the water off his hands and she quickly said, "Sorry."
"Yeah, well, what the fuck," Lucas said. He went back to the refrigerator, took out another bottle of Leinenkugel's and twisted the top off. "Actually, I've been feeling pretty good. Ending it. I'm making some money and I've been out on the road, looking at the world. I was at Little Bighorn a couple of weeks ago. Freaked me out. You can stand by Custer's stone and see the whole fucking fight . . . ."
"Yeah?"
He was marking time, waiting for her to tell him why she'd come to the Cities. But she was better at waiting than he was, and finally he asked, "What're you here for?"
She licked a chip of roast beef from the corner of her mouth, her long tongue catching it expertly. Then: "I want you to come to New York."
"For Bekker?" he asked skeptically. "Bullshit. You guys can handle Bekker. And if I was a New York cop, I'd get pissed off if somebody came in from the outside. A small-town guy."
She was nodding. "Yeah, we can handle Bekker. We've got guys saying all kinds of things: that we'll have him in a week, in ten days . . . . It's been six weeks, Lucas. We'll get him, but the politics are getting ugly."
"Still . . ."
"We want you to jawbone the media. You're good at that, talking to reporters. We want to tell them that we're doing everything we can, that we're even importing the guy who caught him the first time. We want to emphasize that we're pulling out all the stops. Our guys'll understand that, they'll appreciate it—they'll know we're trying to take the heat off."
"That's it? A public-relations trick?" He grimaced, began to shake his head. He didn't want to talk to reporters. He wanted to get somebody by the throat . . . .
"No, no. You'll work the case, all right," she said. She finished the sandwich and held her hands out, fingers spread, looking for a napkin, and he handed her a paper towel. "Right down on the street with the rest of them. And high priority, too. I _do_ value your abilities."
Lucas caught something in her voice. "But?"
"But . . . all of that aside, there's something else."
He laughed. "A third layer? A Lily Rothenburg layer? What're you doing?"
"The thing is, we've got serious trouble. Even bigger than Bekker, if you can imagine it." She hesitated, searching his eyes, intent, then balled up the paper napkin and did a sitting jump shot into a wastebasket before continuing. "This can't come out anywhere."
Irritated, he wordlessly backhanded the comment away, like a bothersome gnat. She nodded, slipped off the stool, took a quick turn around the kitchen, picked up an enamel coffee cup, turned it in her hands, put it down.
"We're looking at thirteen murders," she said finally. "Not Bekker's. Someone else's. These are all . . . hits. Maybe. Of the thirteen—those are the ones we're sure of, we think there are more, as many as forty—ten were out-and-out assholes. Two of them were pretty big: a wholesaler for the Cali cartel and an up-and-coming Mafia guy. The other eight were miscellaneous small-timers."
"Number eleven?"
"A lawyer," Lily said. "A criminal defense lawyer who represented a lot of big dopers. He was good. He put a lot of people back on the street that shouldn't have been there. But most people thought he was straight."
"Hard to be straight, with that job," Lucas said.
"But we think he was. The investigation hasn't turned up anything that'd change our mind. We've been combing his bank records, along with the IRS and the state tax people. There's not a goddamned thing. In fact, there wouldn't have been any point in his being crooked: he was pulling in so much money he didn't need any more. Three million bucks was a slow year."
"Okay. Who was twelve?"
"Number twelve was a professional black . . . spokesman," she said. "A community leader, a loudmouth, a rabble-rouser, whatever you want to call him. But he wasn't a crook. He was a neighborhood politician trying to climb the pole. He was shot in a drive-by, supposedly a couple of gang-bangers. But it was very slick for gang-bangers, good weapons, a stolen car."
"Thirteen?"
"Thirteen was a cop."
"Crooked?"
"Straight. He was investigating the possibility that we've got a rogue group inside the police department, inside intelligence, systematically killing people."
There was a moment of silence as Lucas digested it. "Sonofabitch," he said finally. "They've killed thirteen people for sure, and maybe forty?"
"The cop who was killed—his name was Walter Petty—claimed there were twelve, for sure. He's the thirteenth. We think. He said there could be thirty or forty more."
"Jesus Christ." Lucas pulled at his lip, turned away from her, blankly staring at the microwave. Forty? "You should've picked it up . . . ."
"Not necessarily," Lily said, shaking her head. The short hair whipped around her ears, like a television advertisement, and he caught a smile and suppressed it. This was _business,_ she said. "For one thing, they were killed over a long time. Five years, anyway. And most of them died like you'd have expected, knowing their records. Except more efficiently. That's what you notice when you decide you've got a pattern: the efficiency of it. Bang, bang, they're dead. Never any cops close by—once or twice, they were actually decoyed out. There are never any good witnesses. The getaways are preplanned. No collateral damage, no mushrooms getting knocked down."
"So you've got a pattern of small-time assholes killed by big-time shooters," Lucas said.
"Right. Like this one guy, I met him myself, years ago, when I was just coming off patrol. Arvin Davies." She lifted her eyes to the ceiling and wet her lips, remembering the file. "He was forty-two when he was killed. He was a doper, a drunk. A brawler. He had twenty priors going back to age twelve, and he'd been picked up for one thing or another maybe twenty more times. All small stuff. Street muggings, burglaries, car thefts, rip-offs, possession. He'd get his nose clogged up with angel dust and beat his victims. He killed one five or six years ago, but we could never prove it. He spent twenty years inside, all short time. The last time he got out, he did a couple of muggings and then somebody put him on a wall. Shot him twice in the heart and once in the head. The head shot came when he was already down, a coup de grace. The shooter walked away," she said, hopping back up on the breakfast-bar stool across from him.
"A pro," Lucas said.
"Yeah. And there just wasn't any reason a pro would go after Arvin Davies. He was small-time, chickenshit. But whoever killed him took a real asshole off the streets for good. Maybe forty or fifty nasty crimes a year."
"All the miscellaneous hits are like that?"
"Yup. I mean the techniques are different, but they're all cold, efficient, researched."
Lucas nodded, studying her. "All very enlightening—but where do I come in?"
She looked straight into him, fixing him. "A couple of guys in intelligence spotted the pattern. They got nervous about it. All of the victims, or whatever you'd call them, were heavy in intelligence files. Like the files had been used to choose them. Once they made the report, a secret working group of six ranking officers was set up to monitor it. Petty was eventually brought in to do the dog work."
Lucas interrupted. "He was a shoofly, or whatever you guys call them?"
She shook her head. "He was a crime-scene guy for most of his career, and later on a computer specialist. He was officially a detective second. In this case, he was reporting to the working group under the direct supervision of my boss, John O'Dell. John chairs the working group."
"So there was no past internal-affairs work that might have left a grudge," Lucas said.
"No. And just before he was shot, there was an odd break on the case . . . ." Lily put a hand on top of her head as if she were patting herself, a gesture of thought. "The black guy who was killed, the loudmouth, was named Waites. The file is still open, we still have people digging into it. As a matter of routine, Walt got all the reports coming out of the active cases. He found a report that said a supposed witness to the Waites killing had recognized one of the shooters as a cop. The witness was named Cornell, last name probably Reed. The trouble is, when Walt went looking for him, Cornell Reed had disappeared. Maybe left town. But Walt found him, somehow. He tried to get in touch with us that afternoon, he came by the offices, and when he couldn't, he left a note on voice mail. He said he knew where Reed went."
"Where?"
"We don't know. And Walt was killed that night."
"Jesus—somebody got the voice mail?"
"Unlikely; it's coded," Lily said. "And the shooting was too well set up. They'd planned it ahead of time. If finding the witness had anything to do with it, it was just the trigger that made them go ahead with the shooting."
"Huh. How about Petty's records? Notes?"
"Nothing in his office, but he wasn't keeping anything there, anyway, because of the sensitivity," she said. "He was working out of his apartment, mostly. And that's another thing: somebody got to his apartment before we did. All of his computer disks were gone, and the internal drive—hard drive, is that it?—had been wiped somehow. I don't know how you do it, but there was nothing recoverable."
"Another computer freak?"
"Not necessarily. Whatever they did wasn't fancy. A couple of short commands apparently took care of it. Something like a reformat with a write-over? Does that make sense?"
"Yeah, yeah. Petty must've talked to somebody. It's hard to believe he'd get a break and coincidentally be hit that same night . . . . Who'd he tell about the witness?" Lucas asked.
"We don't know," Lily said. "We _do_ know he came up to our office, after hours, looking for us. O'Dell and I spend a lot of time in a car, going around, putting out political brush fires. We were talking to some people in one of the projects that night. Walt didn't try the car—our driver was waiting in it, and nobody called. The thing is, when Walt came up to the office, he might've bumped into somebody from the working group, there in the hallway. He really wouldn't talk to anybody else, not on this topic."
"So he accidentally bumps into another member of the working group and that guy leaks?"
Lily frowned. "Well . . . the shooting was too quick for a careless leak. Whoever tipped the rogue group did it directly. A phone call. In other words, whoever leaked knows the killers. Maybe he even runs them."
"Sonofabitch. But if you know it's one of the six working-group guys . . ."
Lily shook her head and smiled. "Nothing's ever that easy. For one thing, every one of those six reports to somebody, and they did. And every one of the six has assistants, and some of the assistants know what the working group is doing."
"Doesn't sound very secret," Lucas said.
"Maybe fifteen people know details, and twenty-five know about the problem," Lily said. "That's pretty secret for the department . . . but you see where that leaves us. If one of the working group tipped the killers, he's in a position to know everything. So we're paralyzed. The working group appointed a new lead investigator, an unassigned captain, but he's not doing anything. He's just there to cover our asses, in case something leaks. You know, so we can say we've got an active case under investigation by a ranking officer."
"And you want _me_ to look into it," Lucas said.
Lily nodded. "My boss and I talked it over. We need the work done off the books. Nobody will know but the two of us. It's the only way. And because of Bekker, you're a perfect fit. The goddamn media's going nuts about Bekker, of course, the TV and the _Post_ and _News,_ Doctor Death and all that. You can't get in a cab without hearing a radio talk show about him. So we bring you in, the guy who caught him last time. A consultant. But while you're looking, we're going to put you close to a couple of people Walt was looking at."
"Huh." Lucas sat and thought for a long moment, then he looked up. "This guy who got shot," he said. "You called him Walt, like . . . he wasn't just another guy. Is there something I should know?"
She looked at his face, but not into his eyes: her eyes seemed suddenly blank, as though she were seeing another face. "Walt was my oldest friend," she said.
And she told him about the dream . . . .
_The dream had started the night Petty was killed; it began not with a vision, but with an odor, the smell of ozone, as if electrical circuits were burning somewhere. Then she saw herself, through a haze, but with increasing clarity, seated on a simple marble bench, the kind found in cemeteries, with Petty's bleeding, shattered body stretched across her lap. A_ pietà. _She did nothing at all, but simply sat there, looking into his face. In the dream, the point-of-view closed on the face, like a camera creeping forward, and at the last moment, focused not on an image of Christ-like peace, but on a face that had been shredded by high-velocity slugs, at yellow molars slick with drying blood._ . . .
_A ludicrous image, but one that came, night after night._
_But that wasn't the way it had been, the night Petty was killed._
_Petty's seventy-one-year-old mother had called, confused, incoherent. Her only child had been killed, she said, her voice an ancient moan. Walt was dead, dead . . . Lily could see the old woman in her mind's eye, the narrow gray face bent over the black telephone, body shaking, twitching, the withered hand with the handkerchief, the doilies on_ _the TV behind her, the Sacred Heart on the wall. Lily could even smell it, cabbage and bread dough . . . ._
_The old woman said that Lily had to go to Bellevue to identify Walter. Was there a cop there, Lily asked? Yes, right here, and Father Gomez. And the mayor was coming._
_Lily spoke to the cop. Take care of Gloria Petty, she said, the wife of a cop, the mother of a cop. The last one alive in this family. Then, trembling with fear and grief, she'd gone to Bellevue._
_No pietà at Bellevue._
_Just a body, waxlike, dead, sprawled on a blood-soaked gurney, raw from the pickup. The body was wrapped in layers of plastic, like beef being moved. She noted professionally that one of the slugs had ripped off Petty's cheek, exposing his molars; a preview of Petty as a naked skull, a reminder of Petty's naive, happy smile. The smile that flashed every time he saw her, delighted with her presence._
_She recalled a day from their Brooklyn childhood, when the two of them were seven or eight. Late fall, blue skies, crisp weather, a hint of Halloween. There were maple trees on the block, turning red. She'd been sick and had been kept home from school, but her mother let her out in the afternoon to sit on the stoop._
_And here was Walter, running down the street, a paper held overhead, flapping, joy in his eyes. Her spelling test from the day before. A perfect score. Common enough for Lily, but Walter, so generously pleased for her, that smile, that young blond hair slicked down with Vaseline . . ._
_Come to this, the bloody teeth._
_"That's Walter Petty," she told a tired assistant M.E._
_At home again, changing clothes, preparing herself to see Petty's mother, she thought of her school yearbook. She went into the living room, pulled a box from a built-in cupboard, and found three of them. And his senior picture:_ _his hair never quite right, his face too slender, the slightly dazed smile._
_Lily broke and began to weep. The spasm was uncontrollable, unlike anything she'd experienced before, a storm that ended with dumb exhaustion. Wearily, she finished dressing, started for the door._
_And smelled Petty: Petty in the morgue, the stink of the blood and the body in her nose. She ran back to the bathroom, washed her face and her hands, over and over._
_Early the next morning, after the nightmare interlude with Gloria Petty, as she fought for an hour or two of sleep, she dreamed and saw herself on the marble bench, Walter Petty draped on her lap, broken, torn, his bloody teeth leering from the side of his face._ . . .
_Petty was gone._
"Jesus." Lucas was staring at her. "I didn't know you had . . ."
"What?" She tried to smile. "That kind of depth?"
"That kind of old-time relationship. You know about me and Elle Kruger . . ."
"The nun, yes. What would you do if somebody murdered her?" Lily asked.
"Find whoever did it and kill him," Lucas said quietly.
"Yes," Lily said, nodding, looking straight at him. "That's what I want."
The late-afternoon sun had gone red, then a sullen orange. A heavy atmospheric hush, accompanied by a distant rumbling, announced the line of thunderstorms that Lucas had seen from the roof. When Lily first arrived, Lucas, sitting on the roof, had said, "You're absolutely gorgeous." She'd cooled the sense of contact with a quick, "Don't start, Davenport." But there was an underlying tension between them, and now it sprang up again, riding with them as they moved out of the kitchen, into the living room.
Lily perched on a couch, knees together, fumbled through her purse, found a roll of Certs, tipped a couple of them into her hand, then popped them into her mouth. "You've changed things," she said, looking around the house.
"After Shadow Love, the place was pretty shot up," Lucas said. He dropped onto a leather recliner, sitting on the edge of it, leaning toward her. "Some wiring got wrecked and I needed a new floor. Plaster work. He was shooting that goddamn M-15, it was a mess."
Lily looked away: "That's what they used on Walt. An M-15. A full clip: they emptied a full clip into him. They found pieces of him all over the block."
"Jesus . . ." Lucas groped for something else to say, but all he could find was, "How about you? Are you okay?"
"Oh, sure," she said, and fell silent.
"The last time I saw you, you were on a guilt trip about your old man and the kids . . . ."
"That's not over. The guilt trip. Sometimes I feel so bad I get nauseous," she said.
"Do you see the kids?"
"Not so much," she said sadly, looking away from him. "I tried, but it was wrecking all of us. David was always . . . peering at me. And the boys blame me for leaving."
"Do you want to go back?"
"I don't love him," she said, shaking her head. "I don't even like him very much. I look at him now, and it all seems like bullshit, the stuff that comes out of his mouth. And that's weird, because it used to seem so smart. We'd go to parties and he'd spin up these post-Jungian theories of racism and class struggle, and these phonies would stand around with their heads going up and down like they were bobbing for apples. Then I'd go to work and see a report on some twelve-year-old who shot his mom because he wanted to sell the TV to buy crack, and she wouldn't let him. Then I'd go back home and . . . shit. I couldn't stand listening to him anymore. How can you live with somebody you can't stand listening to?"
"It's hard," he said. "Being a cop makes it worse. I think that's why I spent so much time with Jennifer. She was a professional bullshit artist, but basically, she knew what was what. She spent the time on the streets."
"Yeah . . ."
"So where're you at?" he asked again.
She looked at him unsteadily, not quite nervous, but apprehensive somehow. "I didn't want to get into that right away—I wanted to get you committed first. Will you come?"
"Somebody new?" he asked, his voice light.
"Will you come?"
"Maybe . . . so you've got someone."
"Sort of."
"Sort of? What's that?" He hopped off the chair and took a turn around the room. He wasn't angry, he thought, but he looked angry. He reached down and turned on the TV and a tinny, distant voice instantly cried, "Kirrrbeee Puck-it." He snapped it off again. "What does 'sort of' mean? One foot on the floor at all times? Nothing below the waist?"
Lily laughed and said, "You cheer me up, Davenport. You're so fucking crass . . . ."
"So . . . ?" He went to the window and looked out; the thunderheads were gray, with soaring pink tops, and were bearing down on the line of the river.
She shrugged, looked out the window past him. "So, I was seeing a guy. I still am. We hadn't started looking for an apartment together, but the possibility was out there."
"What happened?"
"He had a heart attack."
Lucas looked at her for a minute, then said, "Why does that make perfect sense?"
She forced a smile. "It's really not very funny, I'm afraid. He's in terrible shape."
"He's a cop?"
"Yeah." The smile faded. "He's like you, in some ways. Not physically—he's tall and thin and white-haired. But he is—was—in intelligence and he loves the streets. He writes articles for the _Times_ op-ed page about the street life. He has the best network of spies in the city. And he has a taste for, mmm . . ." She groped for the right phrase.
"Dark-eyed married women?" Lucas suggested, moving closer.
"Well, that," she said, the tentative smile returning. "But the thing is, he likes to fight . . . did like to fight. Like you. Now he can't walk two dozen steps without stopping for a breath."
"Jesus." Lucas ran a hand through his hair. He'd had nightmares of being crippled. "What's the prognosis?"
"Not so good." Tears glistened at the corners of her dark eyes. At the same moment, she smiled and said, "Shit. I wish I didn't do this." She wiped the tears away with the heel and knuckles of her hand. "This was his third attack. The first one was five years ago. That was bad. The second one was a couple months after the first, and wasn't so bad. Then he was coming back. He'd almost forgotten about them, he was working . . . . Then this third one, this was the worst of all. He's got extensive damage to the heart muscle. And he won't stop working. The doctors tell him to spend a year doing graded exercise, to stay away from work, from the stress. He won't do it. And he's still smoking, I think. He's sneaking them. I can smell them on his clothes . . . in his hair."
"So he's going to die," Lucas said.
"Probably."
"That's not so bad," Lucas said, leaning back, looking at her, his voice flat. "You just say fuck it. You do what you want, and if you go, you go."
"That's what you'd do, isn't it?"
"I hope so," he said.
"Men are such goddamn assholes," Lily said.
After another long silence, Lucas asked, "So what are you doing for sex?"
She started to laugh, but it caught in her throat, and she stood up and picked up her purse. "I better get going. Tell me you'll come to New York."
"Answer the question," Lucas said. Without thinking about it, he moved closer. She noticed it, felt the pressure.
"We're . . . very careful," she said. "He can't get too carried away."
Lucas' chest felt curiously thick, a combination of anger and expectation. The electricity between them crackled, and his voice was suddenly husky. "You never really liked being careful."
"Ah, Jesus, Lucas," she said.
He stepped up to her until he was only inches away. "Push me away," he whispered.
"Lucas . . ."
"Push me away," he said, "I'll go."
She stepped back, dropped her purse. Outside, the first heavy drops of rain careened off the sidewalk, and a woman with a dog on a leash dashed past the house.
She rocked back on her heels, looked down at her purse, then grabbed his shirt sleeve to balance herself, lifted one foot, then the other, pulled off her shoes, and stepped into the hallway that led to the bedroom. Lucas, standing in the living room, watched her go, until halfway down the hallway she turned her head, her dark eyes looking at him, and began to unbutton her blouse.
Their lovemaking, she said later, sometimes resembled a fight, had an edge of violence, a tone of aggression. They might begin with an effort at tenderness, but that would slip and they would be bucking, wrenching, twisting . . . .
That night, as the last of the storm cells rumbled off into Wisconsin, with the room smelling of sweat and sex, she sat on the edge of the bed. She seemed weary, but there was a smile at the corner of her lips.
"I'm such a goddamned slut," she said.
"Oh, God . . ." He laughed.
"Well, it's true," she said, "I can't believe it. I was such a nice girl for so long. But I just _need._ It's not intimacy. You're about as intimate as a fuckin' bear. I need the _sex._ I need to get _jammed._ I really can't believe it."
"Did you know you were going to sleep with me?" Lucas asked. "When you got here?"
She sat unmoving for a moment, then said, "I thought it might happen. So I went to the hotel first, and checked in. In case anyone called."
He ran a fingernail down the bumps of her spine, and she shivered. She was going back to the hotel in case "anyone" called . . . .
"This guy you're sleeping with? 'Anyone'?" Lucas said.
"Yes?"
"What are you going to tell him?"
"Nothing. He doesn't need to know." She turned toward him. "And don't you tell him anything, either, Davenport."
"Why?" Lucas said. "Why would I ever see him?"
"His name's Dick Kennett." In the half-light of the bedroom he could see a tiny, rueful smile lift the corner of her mouth again. "He's running the Bekker case," she said.
## CHAPTER
5
Early morning.
Lucas strolled along Thirty-fifth Street, sucking on half of an orange, taking in the city: looking at faces and display windows, at sleeping bums wrapped in blankets like thrown-away cigars, at the men hustling racks of newly made clothing through the streets.
The citric acid was sharp on his tongue, an antidote for the staleness of a poor night's sleep. Halfway down the block, he stopped in front of a parking garage, stripped out the last of the pulp with his teeth, and dropped the rind into a battered trash barrel.
Midtown South squatted across the street, looking vaguely like a midwestern schoolhouse from the 1950s: blocky, functional, a little tired. Six squad cars were parked diagonally in front of the building, along with a Cushman scooter. Four more squads were double-parked farther up the street. As Lucas paused at the trash basket, disposing of the orange, a gray Plymouth stopped in the street. A lanky white-haired man climbed out of the passenger side, said something to the driver, laughed and pushed the door shut.
He didn't slam the door, Lucas noticed: he gave it a careful push. His eyes came up, checked Lucas, checked him again, and then he turned carefully toward the station. The fingers of his left hand slipped under a brilliant-colored tie, and he unconsciously scratched himself over his heart.
Lucas, dodging traffic, crossed the street and followed the man toward the front doors. Lily had said Kennett was tall and white-haired, and the hand over the heart, the unconscious gesture . . . .
"Are you Dick Kennett?" Lucas asked.
The man turned, eyes cool and watchful. "Yes?" He looked more closely. "Davenport? I thought it might be you . . . . Yeah, Kennett," he said, sticking out his hand.
Kennett was two inches taller than Lucas, but twenty pounds lighter. His hair was slightly long for a cop's, and his beige cotton summer suit fit too well. With his blue eyes, brilliant white teeth against what looked like a lifetime tan, crisp blue-striped oxford-cloth shirt and the outrageous necktie, he looked like a doctor who played scratch golf or good club tennis: thin, intent, serious. But a gray pallor lay beneath the tan, and his eye sockets, normally deep, showed bony knife ridges under paper-thin skin. There were scars below the eyes, the remnants of the short painful cuts a boxer gets in the ring, or a cop picks up in the street—a cop who likes to fight.
"Lily's been telling me about you," Lucas said, as they shook hands.
"All lies," Kennett said, grinning.
"Christ, I hope so," Lucas said. Lucas took in Kennett's tie, a bare-breasted Polynesian woman with another woman in the background. "Nice tie."
"Gauguin," Kennett said, looking down at it, pleased.
"What?"
"Paul Gauguin, the French painter?"
"I didn't know he did neckties," Lucas said uncertainly.
"Yeah, him and Christian Dior, they're like brothers," Kennett said, flashing the grin. Lucas nodded and they went on toward the door, Lucas holding it open. "I fuckin' hate this, people holding doors," Kennett grumbled as he went through.
"Yeah, but when you croak, how'd you like it to say on the stone, 'Died opening a door'?" Lucas asked. Kennett laughed, an easy extroverted laugh, and Lucas liked him for it, and thought: _Watch_ it. Some people could _make_ you like them. It was a talent.
"I could die pulling the tab on a beer can, if they let me drink beer, which they don't," Kennett was saying, suddenly sober. "Hope the fuck it never happens to you. Eat aspirin. Stop eating steak and eggs. Pray for a brain hemorrhage. This heart shit—it turns you into a coward. You walk around listening to it tick, waiting for it to stop. And you're weak. If some asshole mugged me, I'd have to take it."
"I don't want to hear about it," Lucas said.
"I don't want to talk about it, but I do, all the time," Kennett said. "Ready to meet the group?"
"Yeah, yeah . . ."
Lucas followed Kennett through the entrance lobby, waited with him until the reception sergeant buzzed them through to the back. Kennett led the way to a conference room with a piece of notebook paper Scotch-taped to the door: "Kennett Group." The room had four corkboards hung from the walls, covered with notes and call slips, maps of Manhattan, telephones, a couple of long tables and a dozen plastic chairs. In the center of it, a burly, sunburned cop in a white shirt and a thin dog-faced detective in a sport coat were facing each other, both with Styrofoam coffee cups in their hands, voices raised.
" . . . your people'd get off their fuckin' asses, we could get somewhere. That's what's fuckin' us up, nobody wants to go outside because it's too goddamn hot. We know he's using the shit and he's got to get it somewhere."
"Yeah, well I'm not the asshole who told everybody we'd have him in a week, am I? That was fuckin' crazy, Jack. As far as we know, he's buying whatever shit he's using in Jersey, or down in fuckin' Philly. So don't give me no shit . . . ."
A half-dozen more plainclothes cops, in thin short-sleeved shirts and wash pants, weapons clipped to their belts, watched the argument from the plastic chairs spread around the institutional carpet. Four of the six held Styrofoam coffee cups, and two or three were smoking cigarettes, snubbing them out in shallow aluminum ashtrays. One unattended cigarette continued to burn, the foul odor like a fingernail scratch on a blackboard.
"What's going on?" Kennett asked quietly, moving to the front of the room. The argument stopped.
"Discussing strategy," the sunburned cop said shortly.
"Any conclusions?" Kennett asked. He was polite, but pushing. Taking over.
The cop shook his head and turned away. "No."
Lucas found a seat halfway back, the other cops looking at him, openly, carefully, with some distance.
"That's Lucas Davenport, the guy from Minneapolis," Kennett said, almost absently, as Lucas sat down. He'd picked up a manila file with his name on it, and was flipping through memos and call slips. "He's gonna talk to the press this morning, then go out on the street this afternoon. With Fell."
"How come you let this motherfucker Bekker get out?" the sunburned cop asked.
"Wasn't me," Lucas said mildly.
"Should of killed him when you could," dog-face said. Dog-face's two top-middle teeth pointed in slightly different directions and were notably orange.
"I thought about it," Lucas said, staring lazily at dog-face until the other broke his eyes away.
Somebody laughed, and somebody else said, "Shoulda."
Kennett said, "You won't remember this, Davenport, but let me introduce Lieutenants Kuhn, Huerta, White, Diaz, Blake, and Carter, and Detectives Annelli and Case, our serial-killer specialists. You can get the first names sorted out later . . . ."
The cops lifted hands or nodded at him as their names were called out. They looked like Minneapolis cops, Lucas thought. Different names, but the attitude was the same, like a gathering of paranoid shoe salesmen: too little pay, too many years of burgers and fries and Butterfingers, too many people with big feet trying to get into small shoes.
A red-haired woman walked into the room carrying a stack of files, and Kennett added, "And this is Barb Fell . . . . Barb, that's Lucas Davenport in what appears to be a five-hundred-dollar silk-blend jacket and two-hundred-dollar shoes . . . ."
Fell was in her mid-thirties, slender, her red hair just touched with gray. An old scar, shaped like a new moon, cupped one side of her long mouth, a dead-white punctuation mark on a pale oval Welsh face. She sat next to him, perching, shook hands quickly and turned back to the front of the room.
"John O'Dell's coming over, he's going to sit in," one of the cops was telling Kennett. Kennett nodded, dragged a chair around to face the others and said, "Somebody tell me we've got something new."
After a moment of silence, Diaz, a tall, gaunt detective, one of the lieutenants, said, "About the time Bekker would've got here, a cab disappeared. Three months old. One of them new, round Caprices. Poof. Gone. Stolen while the driver was taking a leak. Supposedly."
Kennett's eyebrows went up. "Never seen again?"
"Not as far as we can tell. But, ah . . ."
"What?"
"One of the guys checked around. The driver doesn't know anything from anything. Went into a bar to take a leak, comes out, and it's gone. But the thing had been in two accidents, and the driver says it was a piece of shit. Says the transmission was shot, there was something wrong with the suspension, the front passenger-side door was so tight you could barely open it. I'd bet the sonofabitch is in a river someplace. For the insurance."
Kennett nodded but said, "Push it. We've got nothing else, right?" He looked around. "Nothing from the Laski surveillance . . . ?"
"No. Not a thing," said another of the lieutenants.
"Um . . ." Lucas lifted a finger, and Kennett nodded at him.
"Lily told me about the Laski scam, and I've been thinking about it."
The cops at the front of the room turned in their chairs to look at him. "Like what?" asked Kennett.
"I don't think Bekker'll go for it. He'd think of Laski as a wrong-headed colleague, not somebody he'd hit. Maybe somebody he'd debate. He's an equal, not a subject."
"We got nothing else going for us," snapped Carter, the sunburned cop. "And it's cheap."
"Hey, it's a smart idea," Lucas said. Laski was a Columbia pathologist who had agreed to analyze Bekker's medical papers for the media. He had condemned them, attacked their morality and science, attacked Bekker as a sadist and a psychotic and a scientific moron—all of it calculated to bring Bekker in. Laski, his apartment and his office were covered by a web of plainclothes cops. So far, Bekker hadn't touched any of the trip wires. "That's why I was thinking about it. About variations."
"Like what?" prompted Kennett.
"Back in the Cities, Bekker subscribed to the _Times,_ and I bet he reads it here. If we could set somebody up to give a lecture, some kind of professional speech that would pull him in . . ."
"Don't tease me, darlin'," Kennett said.
"We have some guy lecture on the medical experiments done by Dr. Mengele," Lucas said. "You know, the Nazi dude . . ."
"We know . . ."
"So he lectures on the ethics of using Mengele's studies in research and the ethics of using Bekker's stuff," Lucas said. "And what might come out of their so-called research that's valuable. And we make an announcement in the _Times._ "
The cops all looked at each other, and then Huerta said, "Jesus Christ, man, half the fuckin' town is Jewish. They'd go batshit . . . ."
"Hey, I don't mean any goddamn anti-Semite fruitcake lecture," Lucas said. "I mean some kind of, you know, soft, intellectual, theory thing. I read about this Mengele ethics debate somewhere, so there's something to talk about. I mean, legit. Maybe we get somebody Jewish to front it, so nobody gets pissed off. Somebody with credentials."
"You think that'd do it?" Kennett said. He was interested.
"Bekker couldn't resist, if he heard about it. He's nuts about the topic. Maybe we could arrange for this guy, whoever we get, to have a controversy with Laski. Something that would get in the papers."
Kennett looked at the others. "What do you think?"
Carter tipped his head, grudgingly nodded. "Could you fix it?"
Kennett nodded. "Somebody could. O'Dell, maybe. We could get somebody at the New School. We know Bekker's around there."
"Sounds okay," said Huerta. "But it'll take a while to set up."
"Two or three days," said Kennett. "A week."
"We oughta have him by then . . . ."
"So we cancel. It's like Laski: I don't see any downside, frankly, and it's cheap," Kennett said. He nodded at Lucas. "I'll get it started."
"Quick."
"Yeah," Kennett said. He looked around the room. "All right, so let's go over it. John, what'd we have from Narcotics?"
"We're hassling everybody, but nothing sounds good," said Blake. "Lotsa bullshit, we're chasing it . . ."
As they reviewed the status of the case, and routine assignments, Fell whispered to Lucas, "Your interviews are all set up. A couple of reporters are already here, and three or four more are coming."
Lucas nodded, but as she was about to add something, her eyes shifted away from him toward the door. A fat man walked in, his body swaying side to side, bumping the door frame, small dark eyes poking into the corners of the room, checking off the detectives, pausing at Lucas, pausing at Fell. He looked like H. L. Mencken in the later years. Spidery veins crisscrossed the gray cheeks; his thinning reddish hair was combed straight back with some kind of oil. His jowls were emphasized by a brooding, liverish underlip that seemed fixed in a permanent pout. He wore a three-piece suit in a color that might have been called oxblood, if anyone made oxblood suits.
"O'Dell," Fell said under her breath, at his ear. "Deputy commissioner in charge of cutting throats."
Lily followed O'Dell into the room, picked out Lucas, tipped her head and lifted her eyebrows. She wore a tailored navy-blue suit and a long, mannish red necktie knotted with a loose Windsor. She carried a heavy leather cop's purse over her shoulder, her hand lying casually on the strap at the back of the purse. If she moved her hand four inches, she'd be gripping the butt of a .45. Lucas had seen her use it once, had seen her shove the .45 in a man's face and pull the trigger, the man's face smearing as though he'd been struck with a hammer, all in the space of a tenth of a second . . . .
Lily touched O'Dell's elbow, guided him toward a chair, then moved around where she could sit next to Lucas. "Get a chance to talk to Dick?" she whispered.
"Yeah. He seems like a pretty good guy . . . ."
She looked at him, as though checking to see if he was serious, then nodded and looked away.
O'Dell was up-to-date on the case's progress, and had no particular ideas about what to do next, he told the cops. He just wanted to sit in, to get a feel for the movement. "What about decoys?" he asked. "Somebody downtown suggested that we might put a few people on the street . . . ."
They argued about decoys for a while, a last-resort effort, but Kennett shook his head. "The area's too big," he said. He wandered over to a bulletin board-sized map of Manhattan, ran a finger from Central Park to the financial district. "If he was hitting a specific group, like hookers or gays, then maybe. But there's no connection between the victims. Except some negatives. He doesn't take street people, who'd probably be the easiest . . . ."
"He may specifically pick victims who look healthy," said Case, one of the serial-killer specialists. "This science thing he has—Danny and I think he rules out anybody who's too odd, or diseased or infirm. They'd mess up his findings. The medical examiner reports are all pretty much the same: these people are healthy."
"All right," said Kennett. "So he takes seven people, five female, two male, one black, six white. Two of the whites are Hispanic, but that doesn't seem to mean anything."
"They're all noticeably small, except the first one," Kuhn said. "The second guy was only five-six and skinny."
"Disposal," Huerta grunted.
They all nodded, and there was another long moment of silence, everybody in the room staring at the map of Manhattan.
"It's gotta be a cab," somebody said. "If he can't let anybody see him, and he's gotta have money for drugs, and he's gotta have someplace to gas these people . . . ." One of the cops looked at Lucas: "What are the chances that he had some money stashed? He was pretty well-off, right? Could he have ditched . . . ?"
Lucas was shaking his head. "When we took him, we blindsided him. He thought he was home free. When his wife's estate got into court, all their money was accounted for."
"Okay, that was pretty thin."
"It seems to me that somebody's protecting him," Lucas said. "An old friend or a new friend, but somebody."
Kennett was nodding. "I've worried about that, but if that's right, there isn't much we can do about it."
"We can try pushing his friend, using the media again," Lucas said. "If he depends on somebody else . . ."
O'Dell, seated heavily on a shaky folding chair, interrupted. "Wait, wait. You guys are getting ahead of me. How do we think this, that he has a friend?"
"We've papered the goddamn town with his picture and with simulations of what he'd look like if he dyed his hair or grew a beard or if he shaved his head," said Kennett. "These aren't identikit mock-ups, these are based on good-quality photographs . . . ."
"Yeah, yeah . . ." O'Dell said impatiently.
"So unless he's invisible or living in the sewers, he's probably being protected," Lucas said, picking up the thread from Kennett. "He can't be a regular tenant somewhere. He'd have to pay rent and people'd see him on a regular basis. He can't risk landlords or nosy neighbors."
"And that means he's living with somebody or he's on the street," Kennett said.
"He's not on the street," Lucas said positively. "I can't see him living like that. He just wouldn't do it. He's . . . fastidious. Besides, he's got to have a vehicle. He didn't call a cab to haul these bodies around."
"Unless he drives a cab himself," said Huerta.
"Not much there," said Diaz, shaking his head. "We'll push the stolen one . . . ."
"And it'd still be pretty risky," Lucas said.
"Yeah, but it answers a lot of questions: how he gets transportation, how he makes money and still keeps his face hidden," Kennett said. "If he worked a couple of hours a night, late, and picked his spots . . . maybe concentrated on the tourist and convention areas, you know, the Javits Center, places like that. He'd mostly be dealing with out-of-towners, which would explain Cortese. People trust cabbies. Like if he pretended he had a parcel, gets out and asks somebody where an address was . . ."
"I don't know," said Lucas.
They all stared at the map some more. Too much city; single buildings that would hold the populations of two or three small towns.
"But I still think you might be right, that he's living with somebody," Kennett said finally. "How he gets his money . . ."
"He's got skills," Lucas said. "He's got an M.D., he knows chemistry. A good chemist on the run . . ."
"Methedrine," said White, a bald man in gray knit slacks. "Ecstasy. LSD. It's all back, almost like the old days."
"Be a good reason to protect him, too," said Kuhn. "He'd be a cash cow."
"Assuming this isn't just bullshit, what does it get us?" O'Dell asked impatiently.
"We start looking for ways to put pressure on whoever he's living with or who's covering for him," Lucas said. "We need some heavy-duty contact with the media."
"Why?" said O'Dell.
"Because we have to move them around. Get them to do a little propaganda for us. We need to talk about how anybody who's hiding Bekker is an accessory to mass murder. We need some headlines to that effect. That their only hope is to roll over on him, plead ignorance, get immunity. We've got to chase him out in the open."
"I could call somebody," O'Dell said.
"We need the right emphasis . . ."
"We can figure something out," O'Dell said. "Are you still talking to the reporters this morning?"
"Yeah."
"Throw something in, then . . . ."
When the meeting broke up, O'Dell lurched ponderously out of his chair, leaned toward Lucas, and said, "We'd like to sit in on the press thing. Me and Lily."
Lucas nodded. "Sure." O'Dell nodded and headed toward the front of the room, and Lucas turned to Fell. "We're going out this afternoon?"
"Yeah. They've got us looking for fences," she said. She had gray eyes that matched the touch of gray in her hair; she was five-six or so, with a slightly injured smile and nicotine-stained fingers.
"Could I get copies or printouts of all the Bekker files, or borrow what I can't copy?"
"Right here," she said, patting the stack of manila folders in her lap.
From the front of the room, where he was talking to Kennett, O'Dell called, "Davenport." Lucas stood up and walked over, and O'Dell said, "Dick has been telling me about your idea, the lecture thing, the Mengele. I'll call around this afternoon and set it up. Like for next week. We'll play it like it's been set for a while."
Lucas nodded. "Good."
"I'll see you in the hall," O'Dell said, breaking away. Out of the corner of his eye, as O'Dell spoke to him, Lucas could see Kennett's mouth tic. Disgust? "I've gotta pee."
When he was gone, Lucas looked at Kennett and asked, "Why don't you like him?"
The distaste that had flicked across Kennett's face had been covered in an instant. He looked at Davenport for a long, measured beat and then said, "Because he never does anything but words. Maneuvers. Manipulations. He looks like a pig, but he's not. He's a goddamn spider. If he had a choice between lying and telling the truth, he'd lie because it'd be more interesting. That's why."
"Sounds like a good reason," Lucas said, looking after O'Dell. "Lily seems to like him."
"I can't figure that," Kennett said. They both glanced down the room at Lily, who was talking with Fell. "That pig-spider business, by the way . . . I put my ass in your hands. If he knew I thought that, my next job'd be directing traffic out of a parking garage."
"Not really," said Lucas. Power equations weren't that simple.
Kennett looked at him, amused. "No. Not really. But the asshole could be trouble."
They were both looking toward Lily, and when she tipped her head toward the hall, Lucas started for the door. "You coming?" he asked Fell.
She looked up from one of her files. "Am I invited?"
"Sure. Gotta be careful, though . . . ."
Reporters from three papers and two television stations were waiting, along with two TV cameras. The reporters were in a good mood, joking with him, chatting with each other about problems at the papers. They didn't think much of the story: the interviews were easy and loose, focused on a trap that Lucas had built for Bekker in Minneapolis, and on Bekker himself.
"Really quick," one of the television reporters said to Lucas as the talk was wrapping up, " 'cause we're not going to have much time . . . . You know Michael Bekker. You even visited with him in his home. How would you characterize him? From your personal acquaintance? He's been called an animal . . ."
"To call Bekker an animal is an insult to animals," Lucas said. "Bekker's a monster. That's the only word I can think of that's even close to what he is. He's a real, live horror-show freak."
"Far out," said the reporter, a harried blonde in a uniform blue blazer. She asked her cameraman, "How'd that look?"
"Looked good, that's what they'll use. Let's get a reverse shot on you, reacting . . ."
When the reporters were gone, O'Dell, sitting spread-legged on a folding chair, the way fat men do, nodded approvingly. "That was good. You say Bekker's smart and hard to catch and that everything is being done." His heavy lips moved in and out a couple of times. "Like the blonde broad said, 'Far out.' "
## CHAPTER
6
The Tropic of Sixth Avenue.
The sky was pink from the pollution haze boiling off the asphalt, and heat mirages made the light poles shimmy like belly dancers. Fell pushed the beat-up Plymouth through the cab traffic, one arm out the window, an unfiltered cigarette between her fingers, old-gold rock 'n' roll playing from a personal boombox in the backseat. The Doors, "Light My Fire."
" . . . don't have enough money to fix the air conditioner," she was saying, "but we get three computer terminals so we can do more paperwork, and they're not even new terminals, they're rehabs . . ."
Black and brown arms hung from the driver's windows of the amber taxis beside them, while the paler passengers slumped in back, simmering in their own juices.
"Why fences?" Lucas asked. They were looking for fences. Fell, he'd been told, specialized in burglary and industrial theft, down through the manufacturing district of Manhattan.
"Because Kennett was reading one of these nut-case medical papers Bekker is writing, and figured out that Bekker was taking measurements that you can only take with medical monitoring gear. One of the papers mentions blood pressure taken from a catheter at the radial artery. You gotta have the right stuff . . . ."
"Check the medical-supply houses?"
"Yup, everywhere in North America and the major Japanese and European suppliers. Nothing. Checked the hospitals for stolen stuff and came up empty, but he had to get it somewhere . . . . There are a couple of other guys checking secondary sources . . . ."
They stopped at a traffic light. On the sidewalk, a fruit vendor sat in a plastic lawn chair with a wet rag on his forehead and took a continuous long peel off a red apple, using a thin-bladed stiletto with a pearl handle. A slow-moving, ratty-furred tiger-striped cat walked past him, stopped to look at the dangling peel, then hopped down into the gutter, took a last look around at the daylight world, and dropped into the sewer. Anything to get out of the heat.
" . . . some kind of heat inversion and the temperature never goes down at night, see. That's when things get weird," Fell said, gunning the car through the intersection. "I got a call once where this PR stuck his old lady's head . . ."
"A what?"
"Puerto Rican. Where this Puerto Rican dude stuffed his old lady's head in the toilet and she drowned, and he said he did it because it was so fuckin' hot and she wouldn't shut up . . . ."
They rolled past the Checks Cashed and the Mexican and Indian restaurants, past the delis and the stink of a dog-'n'-kraut stand, past people with red dots on their foreheads and yarmulkes and witty T-shirts that said "No Farting," past bums and sunglassed Mafia wannabes in nine-hundred-dollar loose-kneed suits with shiny lapels.
Past a large woman wearing a T-shirt with a silhouette of a .45 on the front. A newspaper-style map arrow pointed at the gun's muzzle and said, "Official Map of New York City: You Are Here."
"There's Lonnie," Fell said, easing the car to the curb. A taxi behind them honked, but Fell ignored it and got out.
"Hey, whaddafuck . . ."
Fell made a pistol of her thumb and index finger and pointed it at the cabby and pulled the trigger and continued on around the car. Lonnie was sitting on an upturned plastic bottle crate, a Walkman plugged into his ear, head bobbing to whatever sound he was getting. He was looking the other way when Fell walked up and tapped the crate with her toe. Lonnie reared back and looked up, then pulled the plug out of his ear.
"Hey . . ." Lucas turned in front of him, on the other side. Nowhere to run.
"You sold three hundred hypodermic syringes to Al Kunsler on Monday," Fell said. "We want to know where you got them and what else you got. Medical stuff."
"I don't know nothing about that," Lonnie said. He had scars around his eyebrows, and his nose didn't quite line up with the center of his mouth.
"Come on, Lonnie. We know about it, and I don't much give a shit," Fell said impatiently. Her forehead was damp with the heat. "You fuck with us, we take you down. You tell us, we drive away. And believe me, this is something you don't want to get involved in."
"Yeah? What's going on?" He looked like he was about to stand up, but Lucas put his hand on his shoulder, and he settled back on the crate.
"We're looking for this fruitcake Bekker, okay? He's getting medical gear. We're looking for suppliers. You know at least one . . . ."
"I don't know from this Bekker dude," Lonnie said.
"So just tell us where you got them," Lucas said.
Lonnie looked around, as if to see who was watching. "Atlantic City. From some guy in a motel."
"Where'd he get them?" Lucas asked.
"How the fuck would I know? Maybe off the beach."
"Lonnie, Lonnie . . ." said Fell.
"Look, I went to Atlantic City for a little straight action. You know you can't get straight action around here anymore . . . ."
"Yeah, yeah . . ."
" . . . And I meet this guy at the motel and he says he's got some merchandise, and I say, 'Whatcha got?' And he says, 'All sortsa shit.' And he did. He had, like, a million sets of Snap-On tools and some computer TV things and leather flight bags and belts and suits and shit, and these needles."
"What was he driving?" Lucas asked.
"Cadillac."
"New?"
"Naw. Old. Great big fuckin' green one, color of Key Lime pie, with the white roof."
"Think he's still there?"
Lonnie shrugged. "Could be. Looked like he'd been there awhile. I know there was some girls down the way, he was partying with them, they acted like they knew him . . . ."
• • •
They touched a half-dozen other fences, small-time hustlers. At half-hour intervals, Fell would find a pay phone and make a call.
"Nobody home?"
"Nobody home," she said, and they went looking for more fences.
Fell was a cowgirl, Lucas thought, watching her drive. She'd been born out of place, out of time, in the Bronx. She'd fit in the Dakotas or Montana: bony, with wide shoulders and high cheekbones, that frizzy red hair held back from her face with bobby pins. With the scar at the end of her mouth . . .
She'd been jabbed with the broken neck of a beer bottle, she said, back when she was on patrol. "That's what you get when you try to keep assholes from killing each other."
Babe Zalacki might have been a babe once, before her teeth fell out. She shook her head and smiled her toothless pink smile at Lucas: "I don't know from medical shit," she said. "The closest I got to it was, I got three hundred cases of Huggies a couple of weeks ago. Now Huggies, you can sell Huggies. You take them up to Harlem and sell them on the street corners like that . . . ." She snapped her fingers. "But medical shit . . . who knows?"
Back on the street, Fell said, "Sun's going down."
Lucas looked up at the sky, where a dusty sun hung over the west side. "Still hot."
"Wait'll August. August is hot. This is nothin' . . . . Better make a call."
Up the street, a bald man in a jean jacket turned to face a building, braced a hand against it, and began urinating. Lucas watched as he finished, got himself together, and continued down the street. No problem.
Fell came back and said, "He's home. Phone's busy."
They took a half hour, cutting crosstown as the light began to fail, through a warehouse section not far from the water. Fell finally slowed, did a U-turn, and bumped the right-side wheels over the curb. She killed the engine, put her radio on the floor in the backseat, fished a sign out from under the seat and tossed it on the dashboard: "No radio inside."
"Even a cop car?"
"Especially a cop car—cop cars got all kinds of goodies. At least, that's what they think."
Lucas climbed out, stretched, yawned, and ran his thumb along his beltline, under his jacket, until it hit the leather of the Bianchi holster. The street was in deep shadow, with doorway niches and shuttered carports in brick walls. A red brick cube, unmarked by any visible sign or number, loomed overhead like a Looney Tune. Rows of dark windows started three stories up; they were tall and narrow, and from the third to the eleventh floor, dark as onyx. Half of the top floor was lit.
"Lights are on," Fell said.
"Weird place to live," Lucas said, looking around. Scrap paper sidled lazily down the street, borne on a hot humid river breeze. The breeze smelled like the breath of an old man with bad teeth. They were close to the Hudson, somewhere in the twenties.
"Jackie Smith is a weird guy," Fell said. Lucas stepped toward the door, but she caught his arm. "Slow down. Give me a minute." She dug into her purse and came up with a pack of Luckys.
"You've got it bad," Lucas said, watching her. "The habit."
"Yeah, but at least I don't need an alarm clock."
"What?" He stepped into it.
"Every morning at seven o'clock sharp, I wake myself up coughing." When Lucas didn't smile, she peered at him and said, "That was a joke, Davenport."
"Yeah. Inside, I'm laughing myself sick," he said. Then he smiled.
Fell tapped a Lucky on the back of a pack of matches, stuck it in her mouth with a two-finger flipping motion, cupped it with her hands and lit it.
"You're not going to fuck me up, are you?" she asked, her eyes flicking up at him.
"I don't know what that means," Lucas said. He stuck a finger between his collar and his neck. His neck felt like sandpaper. If ring around the collar were a terminal disease, they'd be burying him.
"I saw the pictures of Bekker, after the arrest," Fell said. "He looked like somebody stuck his face in a blender. If you do that in New York, with somebody connected downtown, like Jackie is, your fuckin' _career_ goes in the blender."
"I don't have a career," Lucas said.
"I do," said Fell. "Four more years and I'm out. I'd like to make it."
"What're you going to do when you get out?" Lucas asked, making talk while she smoked. He tipped his head back and looked up again. He seemed to do that in New York, even with buildings only twelve stories tall.
"I'm gonna move to Hollywood, Florida, and get a job as a topless waitress," Fell said.
"What?" She brought him down, startled him.
"Joke, Davenport," she said.
"Right." He looked back up, turning in the street. "Who is this guy?"
She took a drag, coughed, covered her mouth with a rolled fist. "Jackie? He's fairly big. The others we've talked to, they were middle-sized or small-timers. Jackie's a wholesaler. There are three or four of them here in midtown. When somebody hijacks a truck full of Sonys, one of the wholesalers'll get it and parcel it out to the small-timers. If Jackie feels like it, he could put out the word on Bekker to fifty or sixty or a hundred guys. If he feels like it. And those guys could probably talk to a million junkies and thieves. _If_ they feel like it."
"If you know all this . . . ?" He looked at her with a cool curiosity. A man turned the corner behind them, saw them standing on the sidewalk, and went back around the corner out of sight.
"He's got his own business, remaindering stuff," Fell continued. "If somebody has six zillion nuts and no bolts to go with them, he calls up Jackie. Jackie buys them and finds somebody who needs them. That's all legal. If you tag him, you'll find him going in and out of warehouses all day, ten or twenty a day, different ones every day of the week. Talks to all kinds of people. Hundreds of them. Somewhere in the mess, he's got eight or ten people working for him, running the fencing business out the back door of these legit warehouses . . . . It's tough, man. I know he's doing it, but I can't find his dumps."
"He knows you?"
"He knows who I am," she said. "I once sat outside this place for three days, watching who came and went. Running license numbers. It was colder than shit. You know how it gets when it's too cold to snow?"
"Yeah. I'm from . . ."
"Minnesota. Like that," she said, looking down the street, remembering. "So the third night, this guy comes out of the building, knocks on our window, my partner and me, and hands us a Thermos of hot coffee and a couple of turkey sandwiches, courtesy of Jackie Smith."
"Hmph." He looked at her. "You take it?"
"I poured the coffee on the guy's shoes," Fell said. She was talking through her teeth. She took a last drag, grinned at him and flicked the cigarette into the street, where it bounced in a shower of sparks. "The silly shit thought he could buy me with a fuckin' turkey sandwich . . . . C'mon, let's do it."
The warehouse door was built of inch-thick glass poured around stainless-steel rods, with an identical second door six feet farther in. A video camera was mounted on the wall between the two doors. Fell pushed a doorbell marked "Top." A moment later, an electronic voice said, "Yes?"
Fell leaned close to the speaker plate. "Detectives Fell and Davenport to see Jackie Smith."
After a short pause, the voice said, "Step inside and hold your badges in front of the camera."
The door lock buzzed and Fell pulled the door open, and they went inside. Now between the two doors, they held their badges in front of the camera. A second later, the lock on the second door buzzed. "Take the elevator to twelve. It's on the way down," the voice said.
A sterile lobby of yellow-painted concrete block waited behind the second door. There were no windows, only the elevator doors and a steel fire door at the far end of the lobby. The elevators were to the left, and another video camera, mounted in a wire cage near the ceiling, watched them.
"Interesting," Lucas said. "We're in a vault."
"Yeah. You'd have a hell of a time getting this far if Jackie didn't want you in. You'd probably need plastique to do it in a hurry. Then you'd have to get through the fire door, to find the stairs, assuming that the elevator was up and locked. By that time, Jackie'd be gone, of course. I'm sure he's got a bolthole somewhere . . . ."
"And he's probably recording all of this," Lucas said.
Fell shrugged. "I'd like to get him, and I've thought about it—that ain't no secret." Halfway up, she said, "You got a thing with Rothenburg?"
He looked down at her. "Why?"
"Just curious," she said. They watched numbers flickering off the floor counter, and then she said, "When she came in, the way she looked at you, I thought you had a thing."
"Nah . . ."
She shook her head; she didn't believe him. Then the elevator doors opened and they stepped into a lobby identical to the one on the bottom floor: yellow-painted concrete block with a gray steel door set in one wall. Another video camera was mounted in a corner.
"Come in," the disembodied voice said.
The steel door opened on Wonderland.
Lucas followed Fell onto a raised hardwood deck, shaped like a half-moon, overlooking an enormous room. Ten or twelve thousand square feet, Lucas thought, most of it open. Different activity areas were defined by furniture, lights and carpet, instead of walls. The kitchen was to the right; a blond man was peering into a stove, and the odor of fresh hot bread suffused the room. To the left, halfway back, a dark-haired man stood on a square of artificial turf with a golf club.
"Over here," said the voice from the hallway, and the man with the golf club waved at them. Fell led the way, a weaving route through what seemed like an acre of furniture.
A jumble of furniture, with no specific style, Lucas thought: it looked as though it had fallen off the back of a truck. Or trucks—different trucks, from different factories. A king-sized English four-poster bed sat on a huge Oriental carpet, and was covered with an American crazy quilt. A six-foot projection TV faced the bed, and three tripod-mounted video cameras pointed at it.
Behind the TV, a semicircular wall of shoulder-high speakers flanked a conversation pit; a marble-topped table in the center held an array of CD and tape equipment, along with a library of a thousand or more compact discs. The floor beneath the stereo area was hardwood, covered with animal skins: tiger and jaguar, stitched beaver, a buffalo robe, a sleek dark square of what might have been mink. Erroll Garner bubbled out of the speakers, working through "Mambo Carmel."
Beyond the bed, and between the bed and the sports area, a glass shower stall stood out of the floor like an oversized phone booth. Two toilets sat next to it, facing each other, and on the other side, a huge tub.
Smith waited in the sports area, two thirds of the way to the back wall. The wall was pierced by three or four doors. So there were more rooms, Lucas thought . . . .
Smith, his back to them, waggled a driver, drove a golf ball into a net, shook his head, and put the club in a bag that hung from a wall peg. Behind him, a rank of unlit lights waited over what appeared to be a real grass putting green, built on a raised surface. Beyond the green, a stained-glass lamp hung over an antique pool table; and at the back of the room, a basketball net hung from a wall. Below it, a court was complete out to the top of the free-throw circle.
"Can't keep my head down," Smith said. He strode toward them, his golf shoes scuffing over the artificial turf. Smith was a short, barrel-chested, barrel-gutted man with a fuzzy mustache and kinky black hair. He wore a black golf shirt tucked into black pleated slacks, with a woven leather belt circling his waist. A gold chain dangled from his neck, with what looked like a St. Christopher medal. He smiled at Fell and stuck his hand out. "You're the cop who was watching me last year . . ."
Fell ignored the hand. "We need to talk to you about this Bekker guy," she said bluntly. "The guy who's chopping up these people . . ."
"The freak," Smith said. He took his hand back, couldn't find a place for it, and finally stuck it in his slacks pocket. He was puzzled, his mustache quivering. "Why talk to me?"
"He needs money and drugs, and he can't get them legitimately," Lucas said. He'd drifted past the driving area to the putting green. The green's surface was knee high, but dished, to provide a variety of contours. He reached down and pressed his fingers against it. Real grass, carefully groomed, cool and slightly damp to the touch.
"Now that's a hell of a project, right there," Smith said enthusiastically. He picked up a remote control, touched a series of buttons, and the lights over the putting green flickered and came on. "Those are special grow lights," he said, pointing up at the lighting fixture. "Same spectrum as the sun. Joe over there, he knows all about different grasses, he set it up. This is genuine bent grass. It took him a year to get it right."
Smith stepped up and onto the green, walked lightly across it, then turned to look at Lucas. Back to business: "So this guy needs money and drugs?"
"Yeah. And we want you to put the word out on your network. Somebody is dealing with him, and we want him. Now."
Smith picked up a putter that was leaning against the far end. Three balls waited in a rack, and he popped them out, lined up the first one, stroked and missed. The ball rolled past the cup and stopped two feet away.
"Twenty-two feet. Not bad," he said. "When you've got a long lag like that, you just try to get it within two feet of the cup. You pretend you're shooting for a manhole cover. That's the secret to single-bogey golf. Do cops play golf?"
"We need you to put out the word," Fell said.
"Talk into my belly button, said Little Red Riding Hood," Smith said. He lined up another putt, let it go. The ball rolled four feet past the cup. "Fuck it," he said. "Nerves. You guys are putting pressure on me."
"There's no wire," Lucas said quietly. "Neither one of us is wired. We're looking for a little help."
"What do I get out of it?" Smith asked.
"Civic pride," Lucas said. The pitch of his voice had dropped a bit, but Smith pretended not to notice, and lined up the last ball.
"Civic pride? In fuckin' New York?" He snorted, looked up and said, "Excuse the language, Dr. Fell . . . . Anyway, I really don't know what you're talking about, this network."
He walked around the green, squinting at the short putt. The blond man approached with a china platter covered with steaming slices of bread. "Anybody for fresh bread? We've got straight and garlic butter . . . ."
"Fuck the bread," said Fell. She looked at Lucas. "We're not getting to him. Maybe we ought to have the fire department check his . . ."
"Nah, political shit doesn't work with a guy who's really connected," Lucas said. "Mr. Smith sounds like he's connected."
Smith squinted at him. "Who're you? I don't remember you . . . ."
"I've been hired as a consultant here," Lucas said. He wandered back to the driving net, speaking so softly that the others could barely pick up the words. He pulled a three iron out of the golf bag and looked at it. "I used to work in Minneapolis, until I got thrown off the force. I caught Bekker the first time, but not before he killed a good friend of mine. Cut her throat. He let her see it coming. Made her wait for it. Then he sawed right through her neck . . . . She was tied up, couldn't fight back. So later, when I caught Bekker . . ."
"His face got all fucked up," Smith said suddenly.
"That's right," said Lucas. He'd come back, carrying the iron. "His face got all fucked up."
"Wait a minute," said Fell.
Lucas ignored her, hopped up on the putting green, and walked toward Smith. Out of the corner of his eye, he saw Fell's hand sliding into the fold of her shoulder bag. "And I didn't worry about fucking him up. You know why? Because I've got a lot of money of my own and I didn't need the job. I don't need any job."
"What the fuck are you talking . . ." Smith backed away, looked quickly at the blond.
" . . . And Bekker got me really pissed," Lucas said to Smith, his voice riding over the other man's. His eyes were wide, the tendons in his neck straining at his shirt collar. "I mean _really fuckin' pissed._ And I had this pistol, with this big sharp front sight on it, and when I caught him, I pounded his face with the sight until you couldn't tell it was a face. Before that, Bekker'd been really pretty, just like this fuckin' green . . . ."
Lucas pivoted and swung the three iron, a long sweeping swing into the perfect turf. A two-pound divot of dirt and grass sprayed off the platform across the pool table.
"Wait, wait . . ." Smith was waving his hands, trying to stop it.
The blond had set the china tray aside and his hand went toward the small of his back and Fell had a pistol out, pointed at his head, and she was yelling, "No, no, no . . ."
Lucas rolled on, swinging the club like a scythe, screaming, walking around Smith, saliva spraying on Smith's black shirt. "Pounded his face, pounded his motherfuckin' face, you believe the way we pounded his fuckin' face."
When he stopped, breathing hard, a dozen ragged furrows slashed the surface of the green. Lucas turned and looked at the blond man. Hopped down off the platform, walked toward him.
"You were going to pull out a gun," he said.
The blond man shrugged. He had heavy shoulders, like a weight lifter, and he shifted, setting his feet.
"That really pisses me off," Lucas shouted at him.
"Hold it, for Christ's sake," said Fell, her voice low and urgent.
Lucas swung the iron again, quickly, violently, overhead, then down. The blond flinched, but the iron smashed through the freshly baked bread and the platter beneath it. Pieces of china skittered across the floor, and he shouted, "And tried to fuckin' bribe us . . ."
Then he ran down, staggered, turned back to Smith and pointed the club like a saber.
"I don't want to be your friend. I don't want to deal. You're a goddamned dirtbag, and it makes me feel nasty to be here. What I'm telling you is, I want you to put the word out on your network. And I want you to call me. Lucas Davenport. Midtown South. If you don't, I will fuck you up six different ways. I'll talk to the _New York Times_ and I'll talk to the _News_ and I'll talk to _Eye Witness News_ and I'll give them pictures of you and tell them you're working with Bekker. How'd that help business? And I might just come back and fuck you up personally, because this is a serious matter with me, this Bekker thing."
He turned in a half-circle, his breath slowing, took a step toward the door, then suddenly whipped the club into the kitchen like a helicopter blade. It knocked a copper tureen off a wall peg, bounced off the stove, and clattered to the floor with the tureen. "Never was any fucking good with the long irons," he said.
On the way out of the building, Fell watched him until Lucas began to grin.
"Nuttier'n shit, huh?" he said, glancing at her.
"I believed it," she said seriously.
"Thanks for the backup. I don't think blondie would've done much . . . ."
She shook her head. "That was funny; I mean, funny-strange. I didn't know Jackie Smith was gay until I saw this guy. That's like dealing with spouses, only worse. You whack one and the other's liable to come after you with a knife . . . ."
"Are you sure they're gay?"
"Does Raggedy Ann have a cotton crotch?"
"I don't know what that means," Lucas said, laughing.
"It means yes, I'm sure they're gay," she said.
"How come he called you Dr. Fell?" Lucas asked. "Are you a doctor?"
"No. It's from the nursery rhyme: 'I do not love thee, Dr. Fell; the reason why I cannot tell; but this I know, and know full well: I do not love thee, Dr. Fell.' "
"Huh. I'm impressed," Lucas said.
"I know several nursery rhymes," Fell said, digging in her purse for the pack of Luckys. "Want to hear 'Old King Cole'?"
"I mean with Smith. Knowing the rhyme."
"I don't impress you, huh?" She flipped the cigarette into her mouth, her eyes slanting up at him.
"Don't know yet," he said. "Maybe . . ."
Barbara Fell lived on the Upper West Side. They dropped her city car at Midtown South, found a cab, and she said, "I've got a decent neighborhood bar. Why don't you come up and get a drink, chill out, and you can catch a cab from there."
"All right." He nodded. He needed some more time with her.
They went north on Sixth, the sidewalk traffic picking up as they got closer to Central Park, tourists walking arm in arm along the sidewalks.
"It's too big," Lucas said, finally, watching through the window as the city went by. "In the Twin Cities, you can pretty much get a line on every asshole in town. Here . . ." He looked out and shook his head. "Here, you'd never know where it was coming from. You got assholes like other places got raindrops. This is the armpit of the universe."
"Yeah, but it can be pretty nice," she said. "Got the theaters, the art museums . . ."
"When was the last time you went to a theater?"
"I don't know—I really can't afford it. But I mean, if I could."
"Right."
In the front seat, the taxi driver was humming to himself. There was no tune, only variations in volume and intensity as the driver stared blank-eyed through the windshield, bobbing his head to some unheard rhythm. His hands gripped the wheel so tightly that his knuckles were white. Lucas looked at the driver, looked at Fell and shook his head. She laughed, and he grinned and went back to the window.
The bar was small, carefully lit, convivial. The bartender called Fell by her first name, pointed her at a back booth. Lucas took the seat facing the entrance. A waitress came over, looked at him, looked at Fell, said, "Ooo."
Fell said, "Strictly business."
"Ain't it always," the waitress said. "Didja hear Louise had her kid, baby girl, six pounds four ounces?"
Lucas watched Fell as she chatted with the waitress. She looked a little tired, a little lonesome, with that uncertain smile.
"So," she said, coming back to Lucas. "Do you really freeze your ass off in Minnesota? Or is that just . . ."
Small talk, bar talk. A second drink. Lucas waiting for a break, waiting . . . .
Getting it. A slender man walked in, touched a woman on the cheek, got a quick peck in return. He was blond, carefully dressed, and after a moment, looked at the back of Fell's head, said something to the woman he'd touched, then looked carefully at Lucas.
"There's a guy," Lucas said, leaning across the table, talking in a low voice. "And I think he's looking at you. By the bar . . ."
She turned her head and lit up. "Mica," she called. To Lucas she said, "He used to be my hairdresser. He's, like, moved downtown." She slid out of the booth, walked up to the bar. "When did you get back . . . ?"
"I thought that was you . . ." Mica said.
Mica had been to Europe; he started a story. Lucas sipped the beer, lifted his feet to the opposite seat, caught Fell's purse between his ankles, pulled it in. Fumbled with it, out of sight, watching. The waitress glanced his way, lifted her eyebrows. He shook his head. If she came over, if Mica's story ended too soon, if Fell hurried back to get a cigarette . . .
There. Keys. He'd been waiting all day for a shot at them . . . .
He glanced at the key ring in his hand, six keys. Three good candidates. He had a flat plastic box in his pocket that had once held push pins. He'd dumped the pins and filled both the bottom and the lid with a thin layer of modeling clay. He pressed the first key in the clay, turned it, pressed again. Then the second key. The third key he did in the lid; if he made the impressions too close together, the clay tended to distort . . . . He glanced into the box. Good, clean impressions, six of them.
Fell was still talking. He slipped the keys back into her purse, gripped it with his ankles, lifted it back to her seat . . . .
Pulse pounding like an amateur shoplifter's.
Jesus.
Got them.
## CHAPTER
7
Lily called the next morning, "Got them," she said. "We're going to breakfast . . . ."
Lucas called Fell, catching her just before she left her apartment.
"O'Dell called," he said. "He wants me to have breakfast with him. I probably won't make it down until ten o'clock or so."
"All right. I'll run the guy Lonnie told us about, the guy with the Cadillac in Atlantic City. It won't be much . . . ."
"Unless the guy's into medical supplies. Maybe the syringes weren't his only item."
"Yeah . . ." She knew that was bullshit, and Lucas grinned at the telephone.
"Hey, we're driving nails. I'll buy you lunch later on."
The Lakota Hotel was old, but well-kept for New York. It was close to the publishing company that produced Lucas' board games, convenient to restaurants, and had beds that his feet didn't hang off of. From this particular room, he had a view over the roof below into the windows of a glass-sided office building. Not wonderful, but not bad, either. He had two nightstands, a writing table, a chest of drawers, a window seat, a color television with a working remote, and a closet with a light that came on automatically when he opened it.
He went to the closet, pulled out a briefcase and opened it on the bed. Inside was a monocular, a cassette recorder with a phone clip, and a Polaroid Spectra camera with a half-dozen rolls of film. Excellent. He closed the briefcase, made a quick trip to the bathroom, and rode back down to the street. A bellhop, loitering in the phone-booth-sized lobby, said, "Cab, Mr. Davenport?"
"No. I've got a car coming," he said. Outside, he hurried down the street to a breakfast bar, got a pint of orange juice in a wax carton, and went back outside.
After leaving Fell the night before, he'd gone to Lily's apartment and given her the key impressions. Lily knew an intelligence officer who could get them made overnight, discreetly.
"Old friend?" Lucas asked.
"Go home, Lucas," she'd said, pushing him out the door.
And now she called his name again: a black town car slid to the curb, a cluster of antennas sticking out of the trunk lid, and when the back window slid down, he saw her face. "Lucas . . ."
O'Dell's driver was a broad man with a Korean War crew cut, his hair the color of rolled steel. A hatchet nose split basalt eyes, and his lips were dry and thick; a Gila monster's. Lucas got in the passenger seat.
"Avery's?" the driver asked. The front seat was separated from the back by an electric window, which had been run down.
"Yeah," O'Dell said. He was reading the _Times_ editorial page. A pristine copy of the _Wall Street Journal_ lay between his right leg and Lily's left. As he looked over the paper, he asked Lucas, "Did you eat yet?"
"A carton of orange juice."
"We'll get you something solid," O'Dell said. He'd not stopped reading the paper, and the question and comment were perfunctory. After a moment, he muttered, "Morons."
Lily said to the driver, "This is Lucas Davenport next to you, Aaron—Lucas, that's Aaron Copland driving."
"Not the fuckin' piano player, either," Copland said. His eyes went to Lucas. "How are ya?"
"Nice to meet you," Lucas said.
At Avery's, Copland got out first and held the door for O'Dell. Copland had a wide, solid gut, but the easy moves of an athlete. He wore a pistol clipped to his belt, just to the left of his navel, and though his golf shirt covered it, he made no particular attempt to conceal it.
A heavy automatic, Lucas thought. Most of the New York cops he'd seen were carrying ancient .38 Specials, revolvers that looked as though they'd been issued at the turn of the century. Copland, whatever else he might be, was living in the present. He never looked directly at Lucas or Lily or O'Dell as they were getting out of the car, but around them, into the corners and doorways and window wells.
In the closest doorway was a solid oak door with a narrow window at eye height, and below that, a gleaming brass plaque that said AVERY'S. Behind the door was a restaurant full of politicians: they had places like this in Minneapolis and St. Paul, but Lucas had never seen one in New York. It was twenty feet wide, a hundred feet deep, with a long dark mahogany bar to the right side of the entrance. Overhead, wooden racks held hundreds of baseball bats, lying side by side, all of them autographed. A dozen flat Plexiglas cases marched down the left-hand wall opposite the bar, like stations of the cross, and each case held a half-dozen more bats, autographed. Lucas knew most of the names—Ruth, Gehrig, DiMaggio, Maris, Mays, Snider, Mantle. Others, like Nick Etten, Bill Terry, George Stirnweiss, Monte Irvin, rang only faint bells in his memory. At the end of the bar, a double row of booths extended to the back of the restaurant; almost all of the booths were occupied.
"I'll be at the bar," Copland said. He'd looked over the occupants of the restaurant, decided that none of them was a candidate for shooting.
O'Dell led the way back: he was an actor, Lucas realized, rolling slowly down the restaurant like a German tank, nodding into some booths, pointedly ignoring others, the rolled copy of the _Wall Street Journal_ whacking his leg.
"Goddamn town," O'Dell said when he was seated at the booth. He dropped the papers on the seat by his leg. Lily sat opposite him, with Lucas. He peered at Lucas across the table and said, "You know what's happening out there, Davenport? People are stringing razor wire—you see it everywhere now. And broken glass on the tops of walls. Like some goddamned Third World city. New York. Like fuckin' Bangkok." He lowered his voice: "Like these cops, if they're out there. A death squad, like Brazil or Argentina."
A balding waiter with a pickle face came to the table. He wore a neck-to-knees white apron that seemed too neatly blotched with mustard.
"Usual," O'Dell grunted.
Lily glanced at Lucas and said, "Two coffees, two Danish."
The waiter nodded sourly and left.
"You got a reputation as a shooter," O'Dell said.
"I've shot some people," Lucas said. "So has Lily."
"We don't want you to shoot anybody," O'Dell said.
"I'm not an assassin."
"I just wanted you to know," O'Dell said. He groped in his pocket and pulled out a strip of paper and unfolded it. The _Times_ story. "You did a good job yesterday. Modest, you give credit to everyone, you stress how smart Bekker can be. Not bad. They bought it. Have you read the files? On this other thing?"
"I'm starting tonight, at Lily's."
"Any thoughts so far? From what you've seen?" O'Dell pressed.
"I don't see Fell in it."
"Oh?" O'Dell's eyebrows went up. "I can assure you that she is, somehow. Why would you think otherwise?"
"She's just not right. How did you find her?"
"Computer. We ran the dead guys against the cops who busted them. She came up several times. Repeatedly, in a couple of cases. Too many times for it to be a coincidence," O'Dell said.
"Okay. I can see her nominating somebody. I just can't see her setting up a hit. She's not real devious."
"Do you like her?" asked Lily.
"Yeah."
"Will that get in the way?" O'Dell asked.
"No."
O'Dell glanced at Lily and she said, "I don't think it will. Lucas fucks over both men and women impartially."
"Hey, you know I get a little tired . . ." Lucas said irritably.
"Fell looks like another Davenport kill," Lily said. She tried for humor, but there was an edge to it.
"Hey, hey . . ." O'Dell said.
"Look, Lily, you know goddamned well . . ." Lucas said.
"Stop, stop, not in a restaurant," O'Dell said. "Jesus . . ."
"Okay," said Lily. She and Lucas had locked up, and now she broke her eyes away.
The waiter returned with a plate piled with French toast and a small tureen of hot maple syrup. A pat of butter floated on the syrup. He unloaded the French toast in front of O'Dell, and coffee cups in front of Lucas and Lily. O'Dell tucked a napkin into his collar and started on the toast.
"There's something more going on here," O'Dell said, when the waiter had gone. "These three hits we're most worried about, the lawyer, the activist, and Petty himself—I believe these guys may be coming out. The shooters."
"What?" Lucas glanced at Lily, who stared impassively at O'Dell.
"That's my sense, my political sense," O'Dell said. He popped a dripping square of toast into his mouth, chewed, leaned back and watched Lucas with his small eyes. "They're deliberately letting us know that they're out there and that they aren't to be fooled with. The word is getting around. Has been for a couple of months. You hear this shit, 'Robin Hood and his Merry Men,' or 'Batman Strikes Again,' whenever some asshole is taken off. There are a lot of people who'd like the idea that they're out there. Doing what's necessary. Half the people in town would be cheering them on, if they knew."
"And the other half would be in the streets, tearing the place apart," Lily said to Lucas. She turned her head to O'Dell. "There's the other thing, too, with Bekker."
"What?" asked Lucas, looking between them.
"We're told that this is real," she said. She fished in her purse, took out a folded square of paper and handed it to him. A Xerox copy of a letter, addressed to the editor of the _New York Times._
Lucas glanced down at the signature: Bekker. One word, an aristocratic conceit and scrawl.
> . . . taken to task for what I consider absolutely essential experiments into the transcendental nature of Man, and accused of crimes; so be it. I will stand on my intellectual record, and though accused of crimes, as Galileo was, I will, like him, be vindicated by a future generation.
>
> Though accused of crimes, I am innocent, and I will have no truck with criminals. It is in that spirit that I write. On Friday night last, I witnessed an apparent gangland shooting . . . .
"Jesus Christ," Lucas said, looking at Lily. "Was this one of the killings you were talking about?"
"Walt," she said.
Lucas went back to the letter. Bekker had seen the two killers clearly.
> . . . would describe him as white, thick, square-faced with a gray, well-trimmed mustache extending the full length of his upper lip, weighing two hundred and twenty pounds, six feet, two inches tall, sixty-one years old. As a trained forensic pathologist, I would wager that I am not wrong by more than five pounds either way, or by more than an inch in height, or two years in age.
>
> The description of the other one, the one I have called Thin, I will hold to myself, for my own reasons . . . .
"This never ran in the paper?" Lucas asked, looking at O'Dell.
"No. They've agreed to hold it at our request, but they've reserved the right to print it if it seems relevant."
"Do you have any idea who it is? This Thick guy?"
He shook his head: "One of four or five hundred cops—if it's a cop at all."
"You could probably narrow it more than that," Lucas said.
"Not without going public," Lily said. "If we started checking out five hundred cops . . . Christ, the papers would be all over us. But the main thing is, you see . . ."
Lucas picked up her thought: "Bekker can identify two cop killers and he's willing to do it . . . ."
"And for that reason, we think these guys'll make a run at Bekker."
"To shut him up."
"Among other things."
"If they are coming out, they're more likely to go for Bekker," O'Dell said. "They might have to go for him anyway, if they think he can identify two of them. But there's more than that: Killing Bekker would be one way to make their point, that some people have to be killed. Bekker's a nightmare. Who can object to killing him? He's made to order for them, if they can find him."
"This is getting complicated," Lucas said. "I worry about Lily. She's close to this thing, funneling stuff around. What happens if they come after her?"
"They won't," O'Dell said confidently. "Two dead cops would be unacceptable . . . ."
"I'd think one dead cop would be unacceptable."
"One dead cop can be finessed. Denied. Two is a pattern," O'Dell said.
"Besides, I'm not exactly a pushover," Lily said, patting the purse where she kept her .45.
"That'll get your ass killed," Lucas said, anger in his voice. They locked up again. "Anyone's a pushover when the shooters are using a fuckin' machine gun from ambush. You're good, but you ain't bulletproof."
"All right, all right . . ." She rolled her eyes away.
"And there's always Copland," O'Dell said. "When Lily's outside working, she's usually with me in the car. Copland's more than a driver. He's tough as a nail and he knows how to use his gun. I'll have him take her home at night."
"Okay." Lucas looked at Lily again, just for a second, then shifted back to O'Dell. "How'd you get onto Fell? Exactly?"
"Exactly." O'Dell mopped up a river of syrup with a crust of the toast, looked at it for a minute, then popped it in his mouth and chewed, his small eyes nearly closing with the pleasure of it. He swallowed, opened his eyes. Like a frog, Lucas thought. "This is it, exactly. Once or twice a semester I go up to Columbia and lecture on Real Politics, for a friend of mine. Professor. This goes way back. So a few years ago—hell, what am I saying, it was fifteen years ago—he introduced me to a graduate student who was using computerized statistical techniques to analyze voting patterns. Fascinating stuff. I wound up taking classes in statistics, and a couple in computers. I don't look like it"—he spread his arms, as if to display his entire corpulent body—"but I'm a computer jock. When these guys in intelligence found what they thought was a problem, I sorted the killings. There _was_ a pattern. No mistake about it. I called in Petty, who specialized in computer searches and relational work. We turned up almost two hundred possibles. For one reason or another, we eliminated a lot of them and got it down to maybe forty. And twelve of those, we were just about sure of. I think Lily told you that . . ."
"Yeah. Forty. That's a pretty unbelievable number."
O'Dell shrugged. "Some of the killings are probably just what they seem to be—thugs getting killed on the street by other thugs. But not all of them. And I'm sure we missed some. So balancing everything out, I think forty, fifty aren't bad numbers."
"How does Fell fit in?" Lucas asked.
"Petty ran the bad guys against cops who'd know them—a lot of complicated name sorts here, but I've got total access."
"And Fell's name came up . . . ."
"Way too much."
"I hate statistics," Lucas said. "The newspapers were always fuckin' with them back in Minneapolis, drawing stupid conclusions from bad data."
"That's a problem, the data," O'Dell agreed. "We'd certainly never get Fell in court, based on my numbers."
"Mmmph." Lucas looked at Lily and then O'Dell. "I need some heavy time to dig through this . . . ."
"Don't," said O'Dell. He pointed a fork at Lucas' nose. "Your first priority is to find Bekker and to provide a diversion for the media. We need a little air. You've got to do that for real. If this gang is out there, these killers, they won't be easily fooled. Bringing you to New York was supposed to be like bringing in a psychic from Boise: to keep the Boises in the newsroom happy. Everybody's buying it so far. They've got to keep buying it. This other thing has to be way, way in the background."
"What happens if we catch Bekker too soon?" Lucas asked. "Before we identify these guys?"
Lily shrugged. "Then you go home and we find some other way to do it."
"Mmm."
"So. We're in a position where we're hopin' a goddamn psycho holds out for another few weeks and maybe butchers somebody else's kid, so we can run down our own guys," O'Dell mumbled, half talking to himself, staring into the half-eaten sludge pile of toast and syrup. He turned to Lily. "We're really fucked, you know that, Lily? We're really and truly fucked."
"Hey, this is New York," Lucas said.
O'Dell slogged through the rest of the French toast, filling in background on Petty's computer search for the killers.
"Is there any possibility that he turned up something unexpected with the computer?" Lucas asked.
"Not really. Things don't work that way—with a computer, you grind things out, you inch forward. You don't get a printout that says 'Joe Blow Did It.' I think something must have happened with this witness."
When they left the restaurant, O'Dell walked ahead, again nodding into some booths, pointedly ignoring others. Lily grabbed Lucas' sleeve and held him back a step.
"Here." She handed him three keys on a ring.
"That was quick," Lucas said.
"This is New York," she said.
Lucas took a cab from Avery's to Fell's apartment building. The cabdriver was a small man with a white beard, and as soon as Lucas settled in the backseat, he asked, "See _Misérables_?"
"What?"
"Let me tell you, you're missing something," the driver said. He smelled like a raw onion and was soaked with sweat. "Where're you going? Okay—listen, you gotta see _Misérables,_ I mean why d'ya come to New York if you ain't gonna see a show, you know what I mean? Look at the crazy motherfucker over there, you should excuse the language, you think they should let a jerk like that on the streets? Jesus Christ, where'd he learn to drive?" The driver stuck his head out the window, leaning on the horn. "Hey, buddy, where'd you learn to drive, huh? Iowa? Huh? Hey, buddy." Back inside, he said, "I tell you, if the mayor wasn't black . . ."
Lucas called Fell at the office from a pay phone mounted on the outside wall of a parking garage. The garage paint, covered with indecipherable graffiti, was peeling off, to reveal another layer of graffiti. "Barb? Lucas. I gotta run back to my place, just for a minute. Are we still on for lunch?"
"Sure."
"Great. See you in a few minutes," Lucas said. He hung up and looked across the street at Fell's apartment building. A thousand apartments, he thought. Maybe more. Ranks of identical balconies, each with a couple of plants, most with bicycles. Yuppie-cycles, the mountain bikes, in case the riders encountered an off-trail situation in Central Park. Some of them, as high as he could see, were chained to the balcony railings.
The lobby of her building was a glass cage surrounding a guard. At the back were two ranks of stainless-steel mailboxes. The guard, in an ill-fitting gray uniform, was stupidly watchful.
"Where's the sales office?" Lucas asked. A light flickered in the guard's eyes. This situation was specifically covered in his orders. "Second floor, sir, take a right."
"Thanks." Apartment security; it was wonderful, if you had it. Lucas walked back to the elevators, punched two. The second floor had several offices, all down to the right. Lucas ignored them, took a left. Found the stairs, walked up a floor, went back to the elevators and punched sixteen.
The telephone call assured him that Fell was still at Midtown; he didn't have to worry that she'd slipped back home for a snack or to pay bills, or whatever. She lived alone, she'd said. He'd gotten her apartment and home phone numbers from an office roster sheet.
He rode up alone, got out in an empty corridor, took a left, got lost, retraced his steps past the elevators. Her door was green; the others were blue, a tomato-red and beige. Other than that, they were identical. He knocked. No answer. Looked around, knocked again. No answer. He tried a key, hit it the first time, popped the door. The silence inside seemed laced with tension.
_Gotta move, move, move_ . . .
The apartment smelled lightly, inoffensively, of tobacco. The living room had a sliding glass door that led out to the balcony; the doors were covered by off-white curtains, half-drawn. She had a view of a similar building, but if he looked sideways, across the street, Lucas could see another rank of buildings across a gap. The gap was probably the Hudson, with Jersey on the other side.
The apartment was neat, but not compulsively so. Most of the furniture was good, purchased as matched sets. Two green overstuffed La-Z-Boy chairs faced a big color television. A low table sat between the chairs, stacked with magazines. _Elle, Vogue, Guns & Ammo._ More magazines lay on the table, and under it he found a pile of novels. Beside the television was a cabinet with a CD player, a tuner, a tape deck and a VCR. A second table held more magazines, four remote controls, an oversize brandy snifter full of matchbooks—Windows on the World, the Russian Tea Room, the Oak Room, The Four Seasons. They were pristine, and looked as though they'd come from a souvenir packet. Other matchbooks were more worn, half-used—several from the bar they'd visited the night before, one with a crown, one with a chess knight, one with an artist's palette. An ashtray held four cigarette butts.
On the walls around the television were photo portraits: a woman standing on a pier with two older people who might have been her parents, and another picture of the same woman in a wedding veil; a square-shouldered young man on a hillside with a collie and a .22, and another of the young man, grown older, dressed in an army uniform, standing under a sign that said, "I know I'm going to heaven, because I served my time in Hell: Korea, 1952." Something wrong with the young man . . . Lucas looked closer. His upper lip was twisted slightly, as though he'd had a harelip surgically repaired.
Her parents? Almost certainly.
A hallway broke to the left out of the living room. He checked it, found a bathroom and two bedrooms. One bedroom was used as an office and for storage; a small wooden desk and two file cabinets were pushed against one wall, while most of the rest of the space was occupied by cardboard boxes, some open, some taped shut. The other bedroom had a queen-sized bed, unmade, with a sheet tangled by its foot, and two chests of drawers, one with a mirror. An oval braided rug lay underfoot, just at the side of the bed, and a pair of underpants lay in the middle of the rug. A thigh-high woven-bamboo basket with a lid half-hid behind one of the chests. He opened it. Soiled clothes: a hamper.
He could see it. _She sleeps in her underpants, sits up, still tired, yawns, gets out of beds, drops her pants for a shower, figures to toss them in the hamper when she gets back, forgets . . . ._
He went back through the living room to the kitchen, which looked almost unused—a half-dozen water glasses sat in a drying rack in the sink, along with a couple of forks, but no dishes. A Weight Watchers lasagna package lay inside a wastebasket. A bottle of Tanqueray gin sat on the cupboard, two-thirds full. He looked in the refrigerator, found bottles of lime-flavored Perrier and Diet Pepsi, a six-pack of Coors, a bottle of reconstituted lime juice and four bottles of Schweppe's Diet Tonic Water. A sack of nectarines lay on top of the fruit drawer. He touched the stove-top. Dust. A freestanding microwave took up half the counter space. No dust. She didn't cook much.
He did the kitchen first: women hide things in the kitchen or the bedroom. He found a set of dishes, inexpensive, functional. Rudimentary cooking equipment. A drawer full of paper, warranties for all the appliances and electronics in the place. He pulled the drawers out, looked under and behind them. Looked in tins: nothing, not even the flour and sugar that was supposed to be there.
In the bedroom, he looked under the bed and found a rowing machine and dust bunnies the size of wolverines; and in the bedstand drawer, where he found a Colt Lawman with a two-inch barrel, chambered for .38 Specials. Swung out the cylinder: six loaded chambers. He snapped the cylinder back, replaced the weapon as he'd found it.
Looked through the chest of drawers. Bundles of letters and postcards in the top drawer, with cheap jewelry and a sealed box of lubricated Trojans. He looked through the letters, hurrying.
> Dear Barb, Just back from New Hampshire, and you should have come! We had the best time!
>
> Dear Barb: Quick note. I'll be back the 23rd, if everything goes right. Tried to call, but couldn't get you, they said you were out, and I was afraid to wake you during the day. I really need to see you. I think about you all the time. I can't stop. Anyway, see you on the 23rd. Jack.
The letter was in an envelope, and he checked the postmark: four years old. He made a mental note: Jack.
Not much else. He pulled out the drawers. Ah. More paper. Polaroid photos. Barbara Fell, sitting on a man's lap, both holding up bottles of beer. They were naked. She was thin, with small breasts and dark nipples.
He was as thin as she, but muscular, dark-haired, and looked at the camera with a practiced lack of self-consciousness. Another shot: the two of them sitting on what looked like a zebra-skin rug, both nude, their eyes red pinpoints. In the background, a mirror, with a brilliant flash reflecting back at the camera. The camera in the mirror was on a tripod, unattended. No third person. The expression on her face . . . Fear? Excitement? Trepidation?
Another photo, the two of them clothed, standing outside what looked like a police station. A cop? He went back to his briefcase, got the Polaroid out, clipped on the close-up attachment, knelt, and duplicated the photos.
There was nothing else in the bedroom. The bathroom was odorless, freshly scrubbed, but the vanity countertop was a jumble of lipsticks, shampoos, soap, deodorant, a box of something called YeastGard, panty shields, a pack of needles, tweezers, a huge box of Band-Aids and a bottle of sesame body oil. The medicine cabinet held a small selection of over-the-counter items: aspirin, Mycitracin, Nuprin.
He headed for the office.
She was meticulous about her accounts, and everything seemed about right: she had one bank account, a safety-deposit box, and an account with Fidelity Investments, which turned out to be an IRA.
And where was her book? He shuffled through the desk drawers. She must have a personal phone book. She probably carried an annual one with her, but she should have some sort of book she kept at home, that she wouldn't be changing every year. He frowned. Nothing in the desk. He walked out to the front room and looked around the telephone. Nothing there. The phone had a long cord, and he walked over to the pile of magazines on the television table, stirred through them. The book was there, and he flipped it open. Names. Dozens of them. He got the Polaroid and began shooting. When he finished, he'd used all but two shots.
Enough. He looked around, checked the lights and backtracked out of the apartment. The guard was staring stoically at a blank marble wall when Lucas left, and never looked up. The guard's job was to keep people out, not keep people in.
• • •
Kennett and another detective were looking at paper, while a third cop talked on a telephone.
"Barbara's down the hall," Kennett said, looking up when Lucas walked in. "We got you an empty office so you can have a little peace . . . ."
"Thanks," Lucas said.
Fell was sorting through a stack of manila files. He stopped in the doorway, watched her for a moment. She was focused, intent. Attractive. The nude photos popped up in his mind's eye: she looked smaller in the photos, more vulnerable, less vivid. She began paging through a file. After a moment, she felt him in the door, looked up, startled: "Jesus, I didn't hear you," she said.
He stepped inside, walked around the table. Picked up a file: "Robert Garber, 7/12." "Is this everything?"
"Yeah. I've been reading through it. A zillion details," she said. She brushed a lock of hair out of her eyes. "The problem is, we don't need any of it. We know who Bekker is and what he looks like, and he admits in these crazy medical papers that he did the killings. All we have to do is find him; we don't need all the usual shit."
"There must be something . . . ."
"I'll be goddamned if I can see it," Fell said. "The other guys made a list, like the stuff you were talking about at the meeting this morning. He needs an income. He needs a place to hide. He needs a vehicle. He needs to change his face. So they've put out the publicity to employers: watch who you're paying. They've contacted all the hotels and flophouses and anyplace else he might stay. They're talking with the taxi companies, thinking maybe he's moving around in the cab—that would explain how he gases them, using the backseat as a gas chamber. They've gone to all the stores that sell cover-up makeup for people who are disfigured, and every place that sells theatrical makeup. The narcotics guys are talking to dealers, and we're chasing fences. What else is there?"
"I don't know, but it's not enough," Lucas said. He flipped his hand at the stack of paper. "Let's look at the victims first . . . ."
They spent an hour at it. Bekker had killed six people in Manhattan, their bodies found scattered around Midtown, the Village, SoHo and Little Italy. Working on the theory he wouldn't take them far, he was probably south of Central Park, north of the financial district. The zip codes on the envelopes he'd mailed to the medical journals suggested the same thing: three papers, three different zips: 10002, 10003 and 10013.
"He uses halothane?"
"That's what they assume," Fell said, nodding. "They found traces in three people when they were doing the blood chemistry. And that supposedly accounts for the lack of any sign of a struggle. The stuff is quick. Like one-two-three-gone."
"Where did he get it?"
"Don't know yet—we've run all the hospitals in Manhattan, northern Jersey, Connecticut. Nothing yet, but you know, nobody tracks exact amounts of the stuff. You could transfer some from one tank to another. If the tank wasn't gone, how could you tell?"
"Nnn. Okay. But how does he get close enough to whip it on them?" Lucas got up and went out into the hallway, came back with a cone-shaped throwaway water cup. "Stand up."
She stood up. "What?"
He thrust the cup at her face. "If I come at you like this, from the front, I can't get the leverage."
Fell stepped back and the cup came free.
"Even if they got some gas, they could get far enough back to scream," he said.
"We don't know that they didn't scream," said Fell.
"Nobody heard anything."
She nodded. "So if he hits them on the street, he must come up from the back."
"Yeah. He grabs them, pulls them in, claps it over their mouth . . ." He turned her around, clapped the cup over her mouth, his elbow in her spine, his hand hooked over her shoulder. "One, two, three . . . Gone."
"Do it again," she said.
He did it again, but this time, she grabbed his wrist and twisted. The paper cup crumbled and her mouth was open. "Scream," she said. He let go and she said, "That doesn't work too well, either."
"This woman . . . Ellen Foen." Lucas picked up the file, flipped it open. "Statements from her friends say she was very cautious. She'd had some trouble with street people—they hang out in the alley behind the place she worked, going through the dumpsters. She could look out through the glass port in the door while it was still locked, and she always checked before she went out. So if Bekker was there, she must have seen him."
"It was late."
"Nine o'clock. Not quite dark."
"Maybe he was dressed okay. He's not a real big guy—maybe she just wasn't worried."
"But with his face?"
"Makeup. Or . . . I don't know. It makes more sense to me that he's driving a cab. She gets in, he's got one of the security windows between himself and the backseat. He's got it sealed up somehow, and when she shuts the door, he turns on the gas. She passes out. I mean, I just can't see a woman, somebody supposedly cautious, letting a guy get that close to her. And even if he comes up from behind, she'd fight it. You're a hell of a lot bigger than Bekker, but you'd have a hard time holding a mask over my mouth, even from behind."
"Maybe that's why he picks small people, women," Lucas suggested.
"Even so, you just twist away. Even if he gets you, there'd be bruises—but the M.E. hasn't found any bruises. It's gotta be a cab, or something like it."
"But why did Foen take a cab? She was running across the street to get Cokes for everybody. Her boyfriend was supposed to pick her up at nine-thirty, when she got off."
"Maybe . . . fuck, I don't know."
"And look at Cortese. Cortese walks out of this club and across Sixth Avenue, down Fifty-ninth Street toward the Plaza. His friends saw him go in at the Sixth Avenue end. He apparently never arrived at the other end, because there was a phone message for him at the Plaza from nine o'clock on, and he never got it. So he gets picked up on Fifty-ninth between Fifth and Sixth. What happened in there? Why would he flag a cab? He only had to go a few hundred feet."
She shrugged. "I don't know. And it's dark in there, so maybe he got jumped. But you gotta be careful when you start looking for logic, man . . . ."
"I know, I know . . . ."
"It could be anything. Maybe Cortese left his friends because he was looking for a little action."
Lucas shook his head. "He sounds awful straight."
"So does Garber . . . I don't know."
"Keep reading," said Lucas.
She was watching him, he thought. Odd glances, wary. "Is there something wrong?" he asked finally.
After a moment, she asked, "Are you really here working on Bekker?"
"Well . . ." He spread his arms to the stack of paper on the table. "Yeah. Why?"
"Oh, the more I think about it, the odder it seems. We'll catch him, you know."
"Sure, I know," Lucas said. "I'm mostly here for the publicity thing. Take some heat off."
"That doesn't seem quite right either," Fell said. She studied him. "I don't know about you. You hang out with O'Dell. You're not Internal Affairs?"
"What?" He pulled back, surprised. "Jesus, Barbara. No. I'm not Internal Affairs."
"You're sure?"
"Hey. You know what happened to me in Minneapolis?"
"You supposedly beat up somebody. A kid."
"A pimp. He'd cut up a woman with a church key, one of my snitches. Everybody on the street knew about it and I had to do something. So I did. He turned out to be a juvenile—I guess I knew that—and I got hammered by Internal Affairs. There was nothing particularly fair about it. I was just doing what I had to do, and everybody knew it. I got fucked because fucking me was safer than not fucking me. But I'm not Internal Affairs. You can check, easy enough."
"No, no."
She went back to her papers, and Lucas to his, but a minute later he said, "Jesus, Internal Affairs."
"I'm sorry."
"Well . . ."
They took a break, walked two blocks down, bumping hips, and got a booth in a Slice-o'-Pie pizza joint, with gallon-sized paper cups of Diet Pepsi. She liked him: Lucas knew it and let the talk drift toward the personal. He told her about his onetime long-distance relationship with Lily; about the ambiguity now. About his kid.
"I wouldn't mind having a kid," Fell said. "My fuckin' biological alarm clock is banging like Big Ben."
"How old are you?" he asked.
"Thirty-six."
"Any fatherhood prospects on the horizon?"
"Not at the moment," she said. "All I meet are cops and crooks, and I don't want a cop or a crook."
"Hard to meet people?"
"Meeting them isn't the problem. The problem is, the guys I like, don't like me. Eventually. Like five years ago, I was going out with this lawyer dude. Not a big-time lawyer, just a guy. Divorced. Long hair, did a lot of pro bono. And pretty hip. You know."
"Yeah. Exactly. Nice neckties."
"Yeah. He was looking around to get remarried. I mighta. But then one day I was out decoying and this big asshole comes onto me really hard, gets me on a wall, whacks me—he's getting off on whacking me. And I go down and I've got this little hideout piece on my leg, this .25 auto, and he's just bending over to pick me up and I stick the piece in his teeth and his eyes get about the size of dishpans and I back him off, he's saying, 'Hold it, hold it . . .' "
"Where's your backup?"
"They're just running up. They put the guy on the wall and one of them says, 'Jesus, Fell, you're gonna have a mouse bigger'n Mickey'—the asshole'd whacked me right under the eye, right on the eye-socket bone, you know?" She rubbed her eye socket, and Lucas nodded. "Hurt like hell. And I say, 'Yeah?' And they got the guy leaning on the wall with his legs apart, and I say, 'Say good-bye to your nuts, shitbag,' and I punted the sonofabitch so hard his balls had to take a train back from Ohio."
"Yeah?" Lucas laughed. Cop stories were the best stories, and Fell looked positively merry.
"So I tell this story to my lawyer friend and he freaks out. And he's not worried about my eye," she said wryly.
"He's worried about the guy on the wall?"
"No, no. He knew that happened. He didn't mind if _somebody_ did it, he just didn't want me to do it. And I think what really bothered him was my quote: 'Say good-bye to your nuts, shitbag.' I shouldn't have told him that. It really bothered him. I think he wanted to join a country club somewhere, and he could see me sitting out on the flagstone terrace with a mint julep or some fuckin' thing, telling the other country club ladies this, 'Say good-bye to your nuts, shitbag.' "
Lucas shrugged. "You ever tried a cop?"
"Yeah, yeah." She nodded, with a small smile, eyes unfocusing. "A trouser snake. We were hot for a while, but . . . You want a little peace and quiet when you're home. He wanted to go out cruising for dopers."
Lucas took a bite out of a slice of pepperoni, chewed a minute and then said, "A couple of years ago, Lily and I were involved. This is between you and me?"
"Sure." The curiosity was wide on her face, unhidden.
"We were getting intense, this was back in Minneapolis, her marriage was falling apart," Lucas said. "Then this Indian dude shot her right in the chest. Goddamn near killed her."
"I know about that."
"I freaked out. Man. So then we saw each other a few times, but I'm afraid to fly, and she was busy . . . ."
"Yeah, yeah . . ."
"Then last year . . ."
"The actress," Fell said. "The one that Bekker killed."
"I'm like a curse," Lucas said, staring past Fell's head, eyes and voice gone dark. "If I'd been a little smarter, a little quicker . . . Shit."
After lunch, they went back to the paper, working through it, finding nothing. Fell, restless, wandered down to the team room as Lucas continued to read. Kennett brought her back a half-hour later.
"Bellevue," she said, plopping down in the chair across from Lucas.
"What?" Lucas looked at Kennett, leaning in the door.
"Bellevue lost some monitoring equipment from one of its repair shops. We never found out because it wasn't too obvious—everything was accounted for, on paper. But when the stuff didn't come back from repair, somebody checked, and it was gone. The repair people have receipts, they thought it was back on the floor. Anyway, it's been gone for more than a month, and probably more like six or seven weeks. From before the time Bekker killed the first one," Kennett said.
"They lost exactly what Bekker's been using in his papers," Fell said.
"He could've gotten the halothane there, too, and probably any amount of drugs," Lucas said. "All from one source, if it's a staffer."
"Sounds like him," Fell said.
"I'd bet on it," Kennett said. He ran a hand through his hair, straightened his tie. Pissed. "God damn it, we were slow pulling this in."
"What're you going to do?"
"Move very quietly: we don't want to scare anybody off," Kennett said. "We'll start processing Bellevue staffers against criminal records. And we'll touch all the dopers we know, see who knows who on the inside. Then we do interviews. It'll take a few days. Maybe you guys could get back to your fences? See if you could find somebody who handles Bellevue."
"Yeah." Lucas looked at his watch. Almost three. "Let's get back to Jackie Smith," he said to Fell.
Smith met them in Washington Square. The afternoon was oppressively hot, but Smith was cool: he arrived in a gray Mercedes, which he parked by a hydrant.
"I don't want to talk to you. You want to talk to somebody, talk to my lawyer," Smith said as Lucas and Fell walked up. They stood just off the boccie ball courts, under a gingko tree, hiding from the sun.
"Come on, Jackie," Lucas said. "I'm sorry about the goddamn putting green. I got a little overheated."
"Overheated, my ass," Smith snarled. "You know how long it'll take to fix it?"
"Jackie, we really need to make an arrangement, okay?" Lucas said. "Something new came up on this Bekker guy, and you're in a position to help. Like I said last night, it's personal with me. No bullshit. I just need a little information."
"I don't know fuckin' Bekker from any other asshole," Smith said impatiently.
"Hey, we believe you," Lucas said. "And I had to do the green. I had to get your attention—you were blowing us off. Isn't that right?"
Smith stared at him for a long beat, then said, "So what do you want? Exactly?"
"We need the names of guys who can get stuff out of Bellevue."
"That's all you want? Then you'll get off my back?"
"We can't promise," Lucas said. "I can't talk for Barbara—but _I'd_ be a hell of a lot friendlier."
"Jesus Christ, I'm dealing with a fuckin' fruitcake," Smith said. Then: "I don't handle deals at that level. That's small-time."
"I know, I know, but we need a guy who does handle that kind of action. A couple of names, that's all."
"You gonna fuck them over?"
"Not if they talk to me. But if they fuck _me_ over, I'll be back to you."
Fell jumped in with a sales pitch: "Jesus, Jackie, this'd be so easy if you just ride along. It's no skin off your ass. You're actually not helping the cops. You're helping some poor woman who's gonna get her heart cut out, or something."
"Yeah, you're the one who poured my coffee on the street," Smith said, apropos of nothing at all. He looked across the plaza, where a group of black kids were working through a dance routine to rap music from a boombox. "All right," he said. "Two guys. Well: a guy and a woman. They're not actually inside the hospital, but they can put you onto guys who are inside."
"That's all we were asking for . . . ."
"Yeah, yeah. Jesus, you're both full of shit . . . ." Then he started toward his car and said, "I'll be a minute."
"Making a call," Fell said as Smith disappeared into the Mercedes.
He was back in two minutes, with two names and addresses. Lucas wrote the names in his notebook. Smith, with a snort of disgust, turned back to his car, shaking his head.
"Angela Arnold and Thomas Leese," Lucas said to Fell. "Where're these addresses?"
Fell looked and said, "Lower East Side. Never heard of them, though. Want me to run them?"
"Yes. Or just drop them off, get them run overnight," Lucas said, looking at his watch. "Kennett wants to be careful, and I don't want to step on him. Let's not worry about talking to them until tomorrow."
Fell dropped him at the hotel, then went on to Midtown South. Lucas cleaned up, ate dinner in the hotel restaurant, went back to his room and watched the Twins and Yankees through the seventh inning, then caught a cab for Lily's apartment. She buzzed him up and came to the door in her bare feet.
"You're late," she said.
"Got hung up," Lucas said, stepping inside. He'd stayed in her apartment almost two years earlier, when she'd just moved in: the furniture then had a temporary, scrounged look. Boxes had been stacked in the living room, a television had sat on two short metal file cabinets. The kitchen wallpaper had been a bizarre bamboo design, with monkeys; the countertops a well-chipped plastic. Now the place had a careful, colored look: warm rugs over a beige carpet; bright hand-printed graphics on the walls; sparse, but carefully chosen chairs and a broad leather couch. The kitchen was a subtle gold with hardwood counters. He'd stopped by the night before to drop off the key impressions, but hadn't stayed long enough to look around. Now he took a few minutes. "The place looks good," he said finally. He felt a pressure: when he'd been there two years before, they'd spent a lot of time in bed, Lily intent on exploring, feeling, desperate for the intensity of the sex. Now they were polite.
"That's what happens when your marriage splits up. You work on the apartment," she said. She stood close to him, but not too close, one hand just touching the other at her waist, like a hostess. Polite and something else. Wary?
"Yeah, I know."
"I made the back bedroom into an office, everything's stacked up in there. Go on back. Want a beer?"
"Sure." He wandered back to the office, yawned, sat down at the desk, pushed the chair back far enough that he could get his heels on a half-open drawer, picked up the first file. He'd been reading files all day; a million facts floating around free-form.
"Kays, Martin." He flipped the file open. Kays had been arrested twice for rape. Served two years the first time, acquitted the second time. He was suspected in as many as thirty attacks on the Upper West Side. He had had it down to a science, attacking women at night in locked parking garages. He apparently entered when a car exited, ducking under the descending door, then waited until he caught a woman alone in the dark. Half-dozen busts on drug-possession charges, assault, theft, drunkenness.
"Kays," Lily said, looking over his shoulder. "He should've gotten it five years earlier."
"Wrong thinking, _mon capitaine,_ " Lucas said, looking up at her. She handed him a Special Export.
"Yeah, but it's part of the problem: with the exception of the three killings I told you about, including Walt, which they can deny, most people in town would be rootin' for these guys if they knew about them. Especially when they're doing guys like Kays. I doubt we could find a jury that'd convict them."
"You mean it was all right, as long as they were hitting dirtbags?"
"No. Just that if you kill somebody who deserves to die, and will anyway, someday, but maybe fuck up a hundred people's lives before then . . . hurrying the due date along doesn't seem that terrible. Compared to killing innocent people. But these guys aren't hitting criminals anymore, they're attacking . . . freedom."
"I can't operate at that kind of rarefied theoretical level," Lucas said, grinning at her.
"It does sound like wimpy-ass bullshit, doesn't it?" she said.
"It does."
"But it isn't," she said.
"All right."
"If you don't feel it . . . why'd you sign on?" she asked.
He shrugged. " 'Cause you're a good friend of mine."
"Is that enough?"
"Sure. As far as I'm concerned, it's one of the few good reasons for doing anything. I'd hate to kill somebody out of patriotism or duty; I could never be a warden and throw the switch on somebody. But in hot blood, to protect family or friends . . . that's all right."
"Revenge?"
He thought for a minute, then nodded. "Yeah, revenge is in there. I like hunting Bekker. I'm gonna get him."
"You and Barb Fell."
"Yup. Speaking of whom . . ." He dug in his jacket shirt pocket. "Look at these. The guy looks like a cop and she's tight with him, or was." He handed her two of the Polaroids he'd taken at Fell's.
"Oh, Barbara," Lily muttered, looking at them, shaking her head. "I know this guy. Vaguely. He's a lieutenant in Traffic. We'll run him against the killings and see what we get."
"And I've got some names for you. Friends of hers. I don't know how many are cops, but if you could run them . . ."
"Sure."
Lucas stayed until two o'clock, taking notes on a yellow legal pad, when Lily came in and asked, "Find anything?"
"No. And you were right. These guys were the scum of the scum. How many people could put together a list like this?"
"Hundreds," she said. "But Barb Fell was at the intersection of a lot of possibilities."
Lucas nodded, ripped the sheets off the legal pad, folded them and stuck them in his jacket pocket. "I'll keep working her."
Lily's apartment was on the second floor of a converted townhouse. Lucas left at ten after two, the night just beginning to find the soft coolness that lay between the tropical days. He was a little tired, but still awake; at home he might have gone for a walk along the river, smoothing down for bedtime. In New York . . .
The street was reasonably well lit; a taxi loitered in the next block. He turned that way and started walking, hands in his pockets.
There were two of them.
They were big, quick, like professional linebackers.
The cars along the street were parked bumper-to-bumper. The guy behind the Citation got Lucas to turn toward him by dragging something metallic across the bumper, a chilling, ripping sound, like a knife dragged down a washboard.
Lucas instinctively stepped away and half-turned, pivoting toward the sound. Something was happening: a sound like that had to be intentional. His hand dropped to the small of his back, toward the weight of his .45.
And as he turned, the second guy, the guy who'd hidden behind the stoop, charged onto the walk, slashed at Lucas' elbow with a sap, hit him in the spine with a shoulder, and drove him into the Citation.
The pain from the sap was like an explosion, as clear as a star on a cold night, separate from the impact, standing by itself: a skillful, debilitating cop-pain. It began at his elbow and exploded up his arm to his shoulder, and Lucas screamed, thinking he might have been shot, his arm flopping uselessly as he was smashed into the car. He tried to swing the arm back, to clear out to the right, but it wouldn't move.
He saw the other man's hand coming down, and partially blocked it with his left, then was hit in the cheekbone with a fist and rocked back against the car.
The second man, coming over the car's fender, hit him, leather gloves, the second punch in a quick one-two-three combo, and Lucas, back hunched, tried to cover.
Thought: _Clear out, clear out . . ._
He was hit again, across the ear, but this time it didn't hurt: it was stunning and he started down, rolling. A gloved hand struck at him and he grabbed it with his good left hand, pulled it under him, pinned it against his chest, let his weight fall on it. He heard what seemed to be a faraway screaming as they hit the concrete walk, felt a snap; he'd broken something. He felt a dim, distant satisfaction, because he was losing this, they were killing him . . . .
Heard glass breaking, registered it, didn't know what it was, but felt the pressure change.
Thought: _Clear out, clear out._ Let go of the gloved hand, felt it wrench away, and the other man screaming . . . Tried to roll under the car, but it was too close to the curb. Tried to cover his head with his good arm . . .
The .45 was like a thunderbolt.
The muzzle-flash broke over them like lightning, freezing everything in a strobe effect. The attackers wore nylon ski masks and gloves, long-sleeved shirts. The one who'd hit him from behind was pivoting, already running. A sap dangled from his hand, long, leather-bound, with a rounded bulge at the business end. The one whose arm Lucas had broken scrabbled to his feet and screamed, "Jesus . . ." and ran.
The .45 struck down again as Lucas sat down on the curb, his legs gone, trying to roll under the car and away from the lightning, not knowing where it came from, groping in the small of his back with his good arm, but the holster was too far around, trying to free his pistol as the attackers faded like ghosts, without a word, down the sidewalk . . . .
Then silence.
And Lily was there in a cotton nightgown, the .45 in her fist, a ludicrous combination, the soft white human cotton and the dark steel killer Colt.
"Lucas . . ." She maneuvered toward him, controlling the .45, not really looking at him, her eyes searching for targets. "Are you okay?"
"Fuck no," he said.
## CHAPTER
8
Bekker was first astonished, then swept away. When he returned to the bookstore, he glanced at the counterman with a sigh.
"Are you okay?" The counterman was concerned. He had a long neck and a narrow head with small features, like an oversized thumb sticking out of his shoulders. His face was cocked to one side and the store lights glittered off the right lense of his spectacles, lending him a Strangelovian menace.
"I'm fine, I'm fine," Bekker squeaked. He shuffled his feet and looked away, down the store.
The store was fifteen feet wide and forty deep. Vinyl paneling sagged away from the walls behind rough shelving; the linoleum floor was cracked and holed. The narrow aisles smelled of moldy paper, disintegrating bookcovers and the traffic of the unwashed. An obese man stood at a sale table halfway back, under a round antishoplifting mirror, a hardcover Spiderman anthology propped on his gut, feeding a nut-covered ice cream bar into his face. Bekker hadn't even seen him come in.
He looked down at the book in his hands, the book that had taken him away. He'd dug it out of a pile of crap in the Medicine/Anthropology section . . . .
"You didn't move for so long, I thought maybe, I don't know . . ." thumb-face said, his Adam's apple bobbing like a toy boat.
_He's trying to pick me up,_ Bekker thought. The notion was flattering, but unwanted. Nobody was allowed too close. Before the Minneapolis cops had beaten him with their pistols, Bekker had been beautiful, but now Beauty was dead. And though he wore heavy Cover Mark makeup to hide the scars, they were visible in bright light. The _Post_ had carried the pictures, with every cut and scar for the world to see . . . .
Bekker nodded, polite, not speaking, glanced at his watch. He'd been gone five minutes; he must have been an odd sight, a reader frozen, absolutely unmoving, unblinking, for five minutes or more.
_Better leave._ Bekker walked to the counter, head down, and pushed the book across. He'd trained himself to speak as little as possible. Speech could give him away.
"Sixteen-fifteen, with tax," the counterman said. He glanced at the book's cover. "Pretty rough stuff."
Bekker nodded, pushed seventeen dollars across the counter, accepted the change.
"Come back again," thumb-face called, as Bekker went out into the street. The bell above the door tinkled cheerily as he left.
Bekker hurried home, saw his name on the front of a newspaper, and slowed. A picture, a familiar face. What?
He picked up a half-brick that held the newspapers flat. Davenport? Christ, it _was_ Davenport. He snatched up the paper, threw a dollar at the kiosk man and hurried away.
"Want yer change?" The dealer leaned out of the kiosk.
No. He had no time for change. Bekker scuttled down the street, his heels scratching and rapping, trying to read the paper in the dim ambient light. Finally, he stopped in the brilliantly lit doorway of an electronics store, the windows full of cameras, fax machines, tape recorders, calculators, disc players, portable telephones, miniature televisions and Japanese telescopes. He held the paper close to his nose.
> . . . controversial former detective from Minneapolis who is generally credited with solving Bekker's first series of murders and identifying Bekker as the killer. In a fight at the time of the arrest, Bekker's face was badly torn . . .
>
> . . . could have shot him," Davenport said, "but we were trying to take him alive. We knew he had an accomplice, and we believed that the accomplice was dead—but unless we took Bekker alive, we'd never know for sure . . .
Liar. Looking up from the paper, Bekker wanted to scream it: _Liar._ Bekker touched his face, hidden beneath the layers of special cosmetic. Davenport had ripped it. Davenport had destroyed Beauty. Bekker froze, was gone . . . .
A bum came up, saw him in the doorway.
"Hey," the bum said, blocking the sidewalk, and Bekker came back. The bum was not particularly large, but he looked as though he'd been hit often and wasn't afraid to be hit again. Bekker wasn't buying it.
"Fuck off," he bawled, his teeth showing. The bum stepped aside, suddenly afraid, and Bekker went by like a draft of Arctic air. Cursing to himself, Bekker turned the corner, waited for a moment, then stepped back to see if the bum was coming after him. He wasn't. Bekker went on to the Lacey building, muttering, growling, crying. He let himself in the front door, hurried down the basement, dropped into his reading chair.
_Davenport in town._ The fear gripped him for a moment and he flashed back to the trial, Davenport's testimony, the detective staring at him the whole time, challenging him . . . . Bekker lived through the testimony, mind caught, tangled in the random sparking of his mind . . . . And he came back, with a sigh.
What? He had a package on his lap. He looked at it in puzzlement, dumped it. The book. He'd forgotten. _Final Cuts: Torture Through the Ages._ The book was filled with illustrations of racks and stakes, of gibbets and iron maidens. Bekker wasn't interested. Torture was for freaks and perverts and clowns. But near the end of it . . .
_Yes._ A photo taken in the 1880s. A Chinese man, the caption said, had assassinated a prince, and had been condemned to the death of a thousand cuts. The executioners had been slicing him to pieces as the photo was taken.
The dying man's face was radiant.
This was what he'd sought in his own work, and here it was, in a century-old photo. This was the light, the luminance of death, pouring from the face of the Chinese man. It wasn't pain—pain was disfiguring: he knew that from his work. He'd been doing his own photography, but had never achieved anything like this. Perhaps it was the old black-and-white film, something special about it.
Bekker sat and gnawed his thumb, Davenport forgotten, obliterated by the importance of this discovery. Where did the aura come from? The knowledge of death? Of the imminence of it? Was that why old people, at the edge, were often described as radiant? Because they knew the end was there, they could see it, and understood there was no eluding it? Was the knowledge of impending death a critical point? Could that be it? An intellectual function, somehow, or an emotional release, rather than an autonomic one?
Too excited to sit, he dropped the book and took a turn around the room. The matchbox was there, in his pocket; three pills. He gobbled them, then looked at the now empty box. _Here_ was a crisis. He'd have to go back out. He'd been putting it off, but now . . .
He glanced at his watch. Yes. Whitechurch would be working.
He stopped in the bathroom, clumsily fished himself out of the pants, peed, flushed, rearranged himself, then went to the telephone. He knew the number by heart and punched it in. A woman's voice answered.
"Dr. West, please," Bekker said.
"Just a moment, please, I'll page."
A moment later: "West." The voice was cool, New Jersey, and corroded. The voice of a fixer.
"I need some angels," Bekker breathed; he used a breathy voice with Whitechurch.
"Mmm, that's a problem. I'm short. I've got plenty of white, though, and I've got crosses. Almost none of the other," Whitechurch said. He sounded anxious. Bekker was an exceptional customer, white, careful, and paid in cash. A Connecticut schoolteacher maybe, peddling to the kids.
"That's difficult," Bekker said. "How much of the white?"
"I could give you three."
"Three would be good. How many crosses?"
"Thirty? I could do thirty."
"Good. When? Must be soon."
"Make it a half-hour."
"Excellent, half an hour," Bekker breathed, and hung up.
When he'd cleaned out the basement, he'd found a pile of discarded sports equipment—a couple of dried-out leather first baseman's mitts with spiderwebs in the pockets; a half-dozen bats, all badly marred, and one split; a deflated basketball; mold- and dirt-covered baseball shoes with rusted metal spikes; two pairs of sadly abused sneakers; and even a pair of shorts, a tank top and a jock. He'd thrown it all in a long box with a Frisbee, a croquet set and a couple of broken badminton racquets. He'd pushed the box into a dark corner. Anybody looking into it could see all the junk with a glance; nothing good; nothing you'd even want to touch.
Bekker had sliced a C-shaped hatch in the bottom of the basketball and stashed his cash inside. Now he picked up the ball, took out three thousand dollars and carefully put the ball back.
After a quick check in the mirror, he climbed the stairs to the ground floor and padded to the back. Just as he reached the back door, the old woman's voice floated down the stairs. "Alex . . . ?"
Bekker stopped, thought about it, then exhaled in exasperation and walked back across the darkened floor to the staircase. "Yes?"
"I need the special pills." Her voice was shadowy, tentative.
"I'll get them," Bekker said.
He went back down to his apartment, found the brown bottle of morphine, shook two into his hand, and climbed back up the stairs, talking to himself. Images of the deathly radiance played through his mind, and, preoccupied, he nearly stumbled into Bridget Land. Land was standing at the base of the stairs that led up to Edith Lacey's apartment.
"Ah," she said, "I was just leaving, Alex . . . . You have Edie's medicine?"
"Yes, yes . . ." Bekker kept his face turned away, head down, tried to brush past.
"Are the pills illegal? Are they illegal drugs?" Land asked. She had squared herself up to him, her chin lifted, tight, catching his shirt sleeve as he passed her. She had smart, dark eyes that picked at him.
Bekker, his voice straining, nodded and said, "I think so . . . . I get them from a friend of hers. I'm afraid to ask what they are."
"What are you . . ." Land began, but Bekker was climbing the stairs away from her. At the top of the stairs, he glanced back, and Land was turning away, toward the door.
"Please don't tell," Bekker said. "She's in pain . . . ."
"Did you see Bridget?" Mrs. Lacey asked.
"Yes, down below . . . ." He got a glass of water and carried the pills to Mrs. Lacey. She gulped them greedily, hands trembling, smacking her lips in the water.
"Bridget asked me if these were illegal drugs. I'm afraid she might call the police," Bekker said.
Mrs. Lacey was horrified. "You mean . . ."
"They _are_ illegal," Bekker said. "You could never get these in a nursing home."
"Oh no, oh no . . ." The old woman rocked, twisting her gnarled, knobby fingers.
"You should call her. Give her time to get home, and talk to her," Bekker said.
"Yes, yes, I'll call her . . . ."
"Her number's on the emergency pad, by the telephone," Bekker said.
"Yes, yes . . ." She looked up at him, her thin skin papery and creased in the moody light.
"Don't forget . . ."
"No . . ." And then: "I can't find my glasses."
He found them near the kitchen sink; handed them to her without a word. She bobbed her head in thanks and said, "My glasses, my glasses," and shuffled toward the TV. "Have you seen . . . No, you don't watch. I saw Arnold on the news."
Arnold Schwarzenegger. She expected him any day to clean the crooks out of New York.
"I've got to go."
"Yes, yes . . ." She waved him away.
"Call Bridget," Bekker said.
"Yes . . ." From the side, her face glowed blue in the light from the television screen, like a black-light painting. Like the face of the dying Chinese . . .
Ultraviolet.
The idea came from nowhere, but with a force that stopped him at the head of the stairs. Could the illumination of the dying man be related to a shifted spectrum? A light phenomenon that occurred in infrared or ultraviolet, that occasionally strayed into visible light? Was that why some people glowed and others didn't? Was that how an old camera caught it, with the poor, wide-spectrum film of the nineteenth century? He'd seen both ultraviolet and infrared photography as a medical student. Ultraviolet could actually increase the resolution of a microscope, and highlight aspects of a specimen not visible in ordinary light. And infrared could pick up temperature variations, even from dark objects.
But that was all he knew. Could he use his ordinary cameras? How to check?
Excited, excited, the science pounded in his brain. He hurried down the stairs, remembering Bridget Land only at the last minute. He slowed, looked ahead apprehensively, but she was gone.
He hurried out the back, got in the Volkswagen, drove it to the fence, hopped out, unlocked the fence, drove through, checked for intruders, climbed back out, locked the gate behind him. He was flapping, frantic, eager to get on his way, to sustain the insights of the evening.
North across Prince, east across Broadway, keeping to the side streets, the buildings pressing against him, working his way north and east. There. First Avenue. And Bellevue, an aging pile of brick.
Bekker looked at his watch. He was a minute or so early; no problem. He took it slowly, slowly . . . . And there he was, walking toward the bus stop. Bekker leaned across the car and rolled the passenger-side window halfway down, pulled to the curb.
Whitechurch saw him, looked once around, stepped to the window. "Three of the white, thirty crosses, all commercial. Two of the angels, good stuff . . ."
"Only two?" Bekker felt the control slipping, fought to retain it. "Okay. But I'll be calling you in a couple of days."
"I'll have more by then. How many could you handle?"
"Thirty? Could you get me thirty? And thirty more of the crosses?"
"Yeah, I think so," Whitechurch said. "My guy's bringing out a new line. Call me . . . and I'll need twenty-one hundred for tonight."
Bekker nodded, peeled twenty-one one-hundred-dollar bills from the roll in his pocket and handed them to Whitechurch. Whitechurch knew Bekker carried a pistol; in fact, he had sold it to him. Bekker wasn't worried about a rip-off. Whitechurch stuffed the bills in his pocket and dropped a bag onto the front seat.
"Come again," he said, and turned toward the hospital.
Bekker rolled the window up and started back, the sack shoved under the seat; but he knew he wouldn't make it without a sample. He _deserved_ a sample. He'd had a revolutionary idea this night, the recording of the human aura . . . .
He stopped at a traffic light, checked the streets, turned on the dome light and opened the bag. Three fat twists of coke and two small Zip-Loc bags. Thirty small commercial tabs in one, two larger tabs in the other. His hands shook as he kept watch and unrolled one of the twists. Just enough to get home.
The coke jumped him and his head rolled backward with the force of it roaring through his brain like a freight train. After a moment, he started out again, slowly, everything preternaturally clear. If he could hold this . . . His hand groped for the PCP bag, found it; only two. But the coke had him, and he popped them both: the angels would hold the coke in place, build on it . . . . He could see for miles now, through the dark. No problem. His mouth worked, fathering a wad of saliva, and he popped a hit of speed, crunched it in his teeth. Only one, just a sample, a treat . . .
A red light. The light made him angry, and he cursed, drove through it. Another. Even more angry, but he held it this time, rolled to a stop. One more pinch of the white: sure. He _deserved_ one more. One more hit . . .
He hadn't taken an experimental subject in more than a week. Instead, he'd huddled in the basement, typing his papers. He had a backlog now, data that had to be collated, rationalized. But tonight, with the angels in his blood . . . And Davenport in town, looking for him.
In taking the other subjects, he worked out a system: hit them with the stun gun, use the anesthetic. And more important, he'd begun looking for safe hunting grounds. Bellevue was one. There were women around Bellevue all the time, day and night, small enough to handle, healthy, good subjects. And the parking ramp there was virtually open . . . . But Bellevue wasn't for tonight, not after he'd just come from there.
In fact, he shouldn't even think of taking one tonight. He hadn't planned it, hadn't done the reconnaissance that provided his margin of safety. But with the angels in his blood, anything was possible.
A picture popped into his head. Another parking ramp, not Bellevue. A ramp attached to a city government building of some kind.
Parking ramps were good, because they were easy to hide in, people came and went at all hours, many of them were alone. Transportation was easily at hand . . . .
And this one was particularly good: each level of the parking ramp had an entrance into the government building, the doors guarded by combination lockpad. A person entering the ramp in a car would not necessarily walk out past the attendants in the ticket booth. So Bekker could go in, and wait . . . .
The ramp itself had a single elevator that would take patrons to the street. In his mind's eye, he could see himself in the elevator with the selected subject, getting off at the same floor, hitting her with the stun gun as they came out of the elevator, using the gas, hiding her body between a couple of cars, then simply driving around to make the pickup . . . . Simple.
And the ramp was close by, on the edge of Chinatown . . . .
The rational Bekker, trapped in the back of his mind, warned him: _no no no no_ . . .
But Gumball-Bekker cranked the wheel around and headed south, the PCP angels burning in his blood.
Chinatown.
There were people in the street, more than Bekker would have thought likely. He ignored them, the PCP-cocaine cocktail gripping his mind, focusing it: he drove straight into the parking garage, hunched over the wheel, got his ticket, and started around the sequence of up-slanting ramps. Each floor was lit, but he saw no cameras. The sequence had him now, his heart beating like a hammer, his face hot . . . .
He went all the way to the top, parked, opened the cocaine twist, cupped some in his hand, snorted it, licked up the remnant.
And went away . . .
When he came back, he climbed out of the car, taking his collection bag from the backseat. A stairwell wrapped around the elevator shaft and he took the stairs down, quietly, the stairs darker than the main ramp area. Bekker was on his toes, his collecting bag around his shoulder, hand on the stun gun . . . .
At the second floor, he stopped, checked the anesthetic tank and mask. Okay. He rehearsed the sequence in his mind: get behind her, hit her with the stun gun, cover her mouth against screams, ride her down, get the gas. He stepped out of the stairwell, glanced into the tiny elevator lobby. Excellent.
Back to the stairwell.
He waited.
And waited.
Twenty minutes, tension rising. Fished in his pocket, did another cross, chewing it, relishing the bite. He heard a steel door close somewhere overhead, echoing through the ramp, and a few minutes later, a car went down. Then silence again. Five more minutes, ten.
A car came in, stopped on the second floor, high heels on concrete . . . Bekker tensed, his hand going quickly to the tank, flicking the switch once on the stun gun.
Then . . . nothing. The sound of high heels receding. The woman, whoever it was, was walking down the ramp to get out, rather than entering the stairwell or going to the elevator.
Damn. It wasn't working. He glanced at his watch: another ten minutes. No more . . . His mind flashed back to the Twin Cities, to an actress. He'd fooled her by dressing as a gas company employee looking for leaks, had killed her with a hammer. He remembered the impact and the flush . . . . Bekker went away.
And came back, sometime later, with the telltale sigh. At the sound of feet below, and a woman's voice.
The elevator doors opening, one floor below . . .
He picked up his bag, hurried around, went into the lobby, pushed the up button: the Bekker at the back of his mind saying _no no no no,_ the foreground Bekker hot with anticipation . . .
The elevator came up, lurched to a stop, and the doors opened. Inside was a dark-haired woman with an oversized purse, eyes large, one hand in her purse. She hadn't expected the stop at the second floor. She saw Bekker, relaxed. Bekker nodded, stepped inside, waited for the doors to close. The woman had punched six, and Bekker reached for it, then stopped, as if he were also going to six. He stepped against the back of the elevator, looking up at the numbers flashing down at them . . . .
She had a gun in her purse, Bekker thought, a gun or tear gas. He thought about that, thought about that . . . got caught in a loop, thinking about thinking about it . . . and when he came back, groping in his collection bag for the stun gun, they were already at six.
He glanced sideways at the woman, caught her staring at him; he looked away. Eye contact might tell her too much . . . . He glanced again, and the woman seemed to be shrinking away, had her hand in her purse again. A tone sounded, a sharp _bing,_ and the doors slid open. For a moment, neither of them moved, then the woman was out. Bekker followed a few feet behind, turned toward her, slipping his shoes off, expecting to pad after her, catch her unexpectedly . . . .
But the woman suddenly stepped out of her own shoes and began running, and at the same time, looking back at him, screaming, a long, shrill, piercing cry.
_She knew._ . . .
Bekker, frozen for an instant by the scream, went after her, the woman screaming, her purse skidding across the floor, spilling out lipsticks and date books and a bottle of some kind, rolling on the rough concrete . . . . She dodged between two cars, backing toward the outer wall, a can in her hand, screaming . . . .
Tear gas.
Bekker was right behind her, losing his bag, going after her bare-handed, the urgency gripping him, the need to shut her up: _She knows knows knows_ . . .
The woman had braced herself between the cars, her hand extended with the tear gas, her mouth open, her nostrils flexing. No way to get her but straight ahead . . .
Bekker charged, stooping at the last moment, one hand up to block the tear-gas spray. She pressed the can toward him, but nothing happened, just a hiss and the faint smell of apple blossoms . . . .
She'd backed all the way to the ramp wall, the lights of the city behind her, the wall waist-high, her shrill scream in his ears, piercing, wailing.
He went straight in, hit her in the throat with one hand, caught her between the legs with the other, heaved, flipped . . .
And the woman went over the waist-high wall.
Simply went over, as though he'd flipped a sack of fertilizer over the wall.
She dropped, without a sound.
Bekker, astonished at what he'd done, panting like a dog, looked down over the wall as she went. She fell faceup, arms reaching up, and hit on the back of her head and neck.
And she died, like that: like a match going out. From six floors up, Bekker could see she was dead. He turned, looking for someone coming after her, a response to the scream.
Heard nothing but a faraway police siren. Panicked, he ran back to the stairs, up two flights, climbed in the Volkswagen, started it, and rolled down through the ramp. Where were they? On the stairs?
Nobody.
At the exit booth, the woman ticket-taker was standing on the street, looking down at the corner. She came back and entered the booth. She was chewing gum, a frown on her face.
"One-fifty," she said.
He paid. "What's going on?"
"Fight, maybe," she said laconically. "A couple of guys were running . . ."
• • •
Twelve hours later, Bekker hunched over an IBM typewriter, a dark figure, intent, humming to himself "You Light Up My Life," poking the keys with rigid fingers. Overhead, a flock of his spiders floated through the air, dangling from black thread attached to a wire grill. A mobile of spiders . . .
The PCP made the world perfectly clear, and he marveled over the crystal quality of the prose as it poured forth from the machine onto the white paper.
. . . _refuted claims that cerebral-spinal pressure obfuscated reliable intercranial measurements during terminal brain activity as per Delano in TRS Notes [Sept. 86]; Delano overlooked the manifest and indisputable evidence of_ . . .
It simply sang—and that cockroach Delano would undoubtedly lose his job at Stanford when the world saw his professional negligence . . . .
Bekker leaned back, looking up at his spiders, and cackled at the thought. A gumball dropped, and he leaned forward, thoughtful now, Bekker the Thinker. He'd made a mistake this night. The worst he'd made yet. His time was probably coming to an end: he needed more work, he needed another specimen, but he had to be very, very careful.
Mmmm. He turned off the typewriter and laid his manuscript aside, carefully squaring the corners of the paper. Went to the bathroom, washed his face again, stared at the scars. The drugs were still with him, but he was also running down. Might even catch some sleep. When had he last slept? Couldn't remember.
He dropped his clothing on the floor, looked at the clock. Midmorning. Maybe a couple of hours, though . . .
He lay down, listened to his heart.
Closed his eyes.
Almost slept.
But then, just on the edge of oblivion, something stirred. Bekker knew what it was. He felt his heart accelerate, felt the adrenaline spurting into his blood.
He hadn't done her eyes. It had been impossible, of course, but that made no difference. She could see him, the dark-haired woman.
She was coming.
Bekker stuffed a handful of sheet in his mouth, and screamed.
## CHAPTER
9
The car slowed and the window between the front seat and the backseat dropped an inch. The early-morning traffic was light, and they were moving quickly, but O'Dell was grumpy about the early hour. Lily hadn't slept at all.
"You want a _Times_?" Copland asked over his shoulder.
"Yes." O'Dell nodded, and Copland eased the car toward the curb, where a vendor waved newspapers at passing cars. A talk show babbled from the front-seat radio: Bekker and more Bekker. When Copland rolled his window down, they could hear the same show from the vendor's radio. The vendor handed Copland a paper, took a five-dollar bill, and dug for change.
"I'm worried," Lily said. "They could try again."
"Won't happen. They didn't mean to kill him, and coming after him again, that way, would be too risky. Especially if he's this tough guy you keep telling me about . . . ."
"We thought they wouldn't go after him the first time . . . ."
"We never thought they'd try to mug him . . . ."
Copland handed a copy of the _Times_ into the backseat. A headline just below the fold said, "Army Suspects Bekker of Vietnam Murders."
"This has gotta be bullshit," O'Dell grumbled, scanning the story. "Anything from Minneapolis?"
"No."
"Dammit. Why don't these assholes check on him? For all they know, the Minneapolis story could be a cover for an Internal Affairs geek."
"Not a thing, so far. And the people in Minneapolis are looking for it."
Silence, the car rolling like an armored ghost through Manhattan.
Then: "It must be Fell. It has to be."
Lily shook her head: "Nothing on her line. She got one call, from an automated computer place saying that she'd won a prize if she'd go out to some Jersey condominium complex to pick it up. Nothing on the office phone."
"Dammit. She must be calling from a public phone. We might need some surveillance here."
"I'd wait on that. She's been on the street for a while. She'd pick it up, sooner or later."
"Had to be Fell, though. Unless it really was muggers."
"It wasn't muggers. Lucas thinks they were cops. He says one of them was carrying a black leather-wrapped keychain sap; about the only place you can buy them is a commercial police-supply house. And he says they never went for his billfold."
"But they weren't trying to kill him."
"No. But he thinks they were trying to put him out of commission. Maybe break a few bones . . ."
"Huh." O'Dell grunted through a thin smile. "You know, there was once a gang on the Lower East Side, they'd contract to bite a guy's ear off for ten bucks?"
"I didn't know that," said Lily.
"It's true, though . . . . All right. Well. With Davenport. String him along . . . ."
"I still feel like I'm betraying him," Lily said, looking away from O'Dell, out the window. A kid was pushing a bike with a flat tire down the sidewalk. He turned as the big black car passed, and looked straight at Lily with the flat gray serpent's eyes of a ten-year-old psychopath.
"He knew what he was getting into."
"Not really," she said, turning away from the kid's trailing eyes. She looked at O'Dell. "He thought he did, but he's basically from a small town. He's not from here. He really doesn't know, not the way we do . . . ."
"What'd you tell Kennett, about why Davenport was at your place?"
"I . . . prevaricated," Lily said. "And I could use a little backup from you."
"Ah."
Lucas hadn't been badly hurt, so Lily flagged a cab, took him to Beth Israel, then reported the attack. Because she'd fired her weapon, there had been forms to fill out. She'd started that night, and called Kennett to tell him about it.
"Should I ask why he was at your place at two in the morning?" Kennett had asked. He'd sounded amused, but he wasn't.
"Um, you don't want to know," Lily had said. "But it was strictly business, not pleasure."
"And I don't want to know."
"That's right."
After a moment: "Okay. Are you all right? I mean, really all right."
"Sure. I've got a busted window I've gotta get fixed . . . ."
"Good. Get some sleep. I'll talk to you tonight."
"That's all? I mean . . . ?"
"Do I trust you? Of course. See you tonight."
Lily looked out the car window, at the city rolling past. Maybe she was betraying Lucas. Maybe she was betraying Kennett. She wasn't sure anymore.
O'Dell said, "Cretins," and his paper shook with anger.
## CHAPTER
10
The reporters came and went, the naive ones swallowing Lucas' story that he had been mugged, others not so sure. A reporter from _Newsday_ said flatly that something else was going on: that Bekker had a gang, or that somebody else was trying to stop Lucas' investigation.
"I don't know about muggers in Minneapolis, but in New York they don't work in professional tag teams. Unless you're lying, you were done by professionals . . . ."
After they were gone, Lucas took a few more Tylenol, wandered down to the bathroom and got back in time to see Lily coming down the hall.
"You look . . . pretty rough," she said.
"It's my cheek. My cheek hurts like hell," he said. He touched a swollen magenta bruise with his middle finger. "At least the headache's going away. They're letting me out after lunch."
"I heard," Lily said.
"Thanks for sending the jeans over. The other pants . . ."
"Are shot." Lily said.
"Yeah."
"O'Dell's fixed the Mengele speech—there'll be a notice in all the papers this afternoon, the _Times_ tomorrow morning, and we're asking everybody to do a note about it. TV, too. We found a guy, a legit guy, who already lectures on Mengele."
"Terrific," Lucas said. "When?"
"Monday."
"Jesus, that quick?"
"We gotta do everything quick. Maybe we can get him before he does another one . . . ." Lily backed into a hospital chair, dropped her purse by her foot. "Listen, about last night. Are you absolutely sure they were cops?"
"Fairly sure. They could have been professional bone-breakers, but it didn't feel that way. They felt like cops. Why?"
"I was thinking about another possibility."
"Smith?"
"Yeah. After you chopped up his putting green . . ."
Lucas pulled his lip. "Maybe," he said. "But I doubt it. One thing you learn as a sleazoid businessman is to roll with the punches."
"Have you talked to Fell this morning?"
"She's on her way over. We have a line on a couple of people who might know something about Bellevue. She's been talking to Kennett, to make sure we don't step on any toes . . . ."
"Okay. I've got Bobby Rich coming over. He's the guy who took the tip about the witness."
"The witness Petty found . . ."
"Yeah, the day he got killed. And there's still some more paper to look at."
"That's pointless, I think," Lucas said. "With these guys, the dead guys, we won't find anything in their lives that'll point to the killers. It has to be bureaucratic: who pulled their files, and when . . ."
"That's impossible."
"Yeah, I know."
"So we're stuck?"
"Not quite, but it's getting sparse. Maybe Rich'll have something. We've still got Fell. I want to take a look at Petty's apartment, his personal stuff. And I wouldn't mind seeing the place where he was shot."
"That's about a half-mile from my apartment—we could walk. His apartment's sealed. I'll get some new seals and take you over. When?"
"Tonight? After we talk to Rich?"
"Fine."
"What'd you tell Kennett . . . ?"
"About you being at my apartment? I said you came over to visit. I told him that sex was not a consideration, last night or in the future. I told him that you weren't making any moves on me and I wasn't making any moves on you, but that we had things to talk about."
"Sounds pretty awful," Lucas said, grinning.
"It could have been, but I just came out with it. I also told him O'Dell was there part of the time. John will back that up."
A few minutes later, Kennett and Fell arrived together, and Lily blew up: "For Christ's sakes, Dick, what're you doing here? Did you walk all the way in?" Hands on hips, she turned to Fell, angry. "Barbara, did you let him . . . ?"
"Shut up, Lily," Kennett said. He touched her cheek with an index finger. To Lucas, he said, "Well, you look like shit."
"What do you think, Barb?" Lucas asked.
Fell had taken cover behind Kennett, and she peered out and said, "He's right. You look like shit."
"Then it's unanimous," Lucas said. "That's what Lily said when she came in. The only one who didn't was a twenty-four-year-old _Times_ reporter with a great ass, who thought I looked pretty good and would probably like to hear more about this case from the hero of it . . . ."
"Gotta be a concussion," Fell said to Lily.
"He's always been like this," Lily said. "I think it's native stupidity."
Kennett, shaking his head, said, "Goddamn women, they're always impressed by a beat-up face. I used to get beat up whenever I needed to get laid. Worked like a charm . . ." He stopped, and frowned at Lucas: "Are you trying to get laid?" and his eyes flicked sideways at Lily.
Fell said, "Not very hard."
Lucas and Kennett laughed; Lily didn't.
Kennett said, "Listen, I wanted to tell you. Go ahead with those names you got. Barb's run them down . . . ."
"One good address and one probable," Fell said.
"Junkies?"
"Nope. Neither one of them. Not the last anybody heard, anyway."
"All right." Lucas eased down from the hospital bed. "Let's go down to the nursing station. Maybe I can talk my way out before lunch."
The charge nurse said the attending physician wanted another look at him: she'd send him down as soon as he arrived, which should be within the next few minutes. "We'll see you first," she said.
"All right, but pretty quick?"
"Soon as he gets here."
Lily said, "I've gotta go. Take it easy today."
"Yeah."
He walked gingerly back to the room with Fell, trying not to move his head too quickly. At the door, he looked back toward the elevators. Kennett and Lily were waiting, looking up at the numbers above the door, then Kennett leaned toward Lily, and she went up on her toes, a kiss that wasn't taken lightly by either of them. Lucas turned away, and caught Fell watching him watch Lily and Kennett.
"True love," he said wryly.
The hot, hazy sun left him feeling faintly nauseous, and the headache lurked at the back of his skull.
"You look pale and wan," Fell said.
"I'm all right." He looked up at the storefront: Arnold's TV and Appliance, Parts & Repair. "C'mon, let's do her."
A bell tingled when they went through the door; a heavyset woman looked up from a ledger, slapped it shut, and moved ponderously to the counter. "Can I help ya?" She had a cheerfully yellow smile and an improbable West Virginia hills accent. To Lucas: "Whoa, you look like you've been in a dustup."
"We're police officers," Fell said. She lifted the flap of her purse, flashed the badge. "Are you Rose Arnold?"
The woman's smile sagged into a frown. "Yeah. What'd you want?"
"We're looking for a guy," Lucas said. "We thought you could help."
"I ain't been here all that long . . . ."
Lucas dug in his pocket, took out his money clip, freed his driver's license and handed it to Arnold. "Barbara here"—he nodded at Fell—"is a New York cop. I'm not. I'm from Minneapolis. They brought me in to help look for this Bekker dude who's chopping people up."
"Yeh?" Arnold was giving nothing away, watching him with her small wandering eyes like a pullet who suspects the axe.
"Yeah. He killed my woman out there. Maybe you read about it. I'm gonna catch him and I'm gonna do him."
Arnold nodded and asked, "So what's that got to do with me?"
"We think he's getting stuff—drugs and medical equipment—from Bellevue. We know that you handle stuff out of Bellevue."
"That's bullshit, I never touch nothing . . . ."
"You moved five hundred cases of white Hammermill Bond copy paper out of there two weeks ago, paid a dollar a case, and sold it to a computer supply place for three dollars a case," Fell said. "We could bust you if we wanted to, but we don't want to. We just want some help."
She looked at them, quietly, a gleam of strong intelligence in her eyes. Calculating. Lucas had a quick vision of her jerking some crappy piece of hillbilly iron out of a drawer, something like a rusty Iver Johnson .32, and popping him in the chest. But nothing happened, except the sound of flies bumping against the front window.
"Killed your woman?" she asked. She tipped her head, looking at him from the corner of her eyes.
"Yeah," he said. "It's real personal."
She mulled it over for another few seconds, then asked, "What do you want?"
"I need the name of a guy who rips stuff out of there on a regular basis."
"Will this come back on me?"
"No way."
She thought about it, then mumbled: "Lew Whitechurch."
"Lew . . ."
"Whitechurch," she said.
"Who else?"
"He's the only one, right out of Bellevue . . . ."
"Any chance he might be peddling pills, too?"
"I think he might. I never touch them, but Lew . . . he's got a problem. He takes a little nose."
"Thanks," Lucas said. He took a personal business card from his pocket, turned it over, wrote his hotel phone number on it. "Have you handled, or know anybody who has handled, a load of emergency-room monitoring equipment?"
"No." Her voice was positive.
"Ask around. If you find somebody, have them call me. It'll never get past us, I swear it on a Bible. I'm only in this because Bekker cut my woman's throat."
"Cut her throat?" The fat woman touched her neck.
"With a bread knife," Lucas said. He let the bitterness flow into his voice. "Listen: anybody dealing with Bekker is liable to find himself strapped to an operating table, eyelids cut off, getting his heart sliced out while he's still alive . . . . You read the papers."
"Watch TV." She nodded.
"Then you know."
"Fuckin' lunatic, is what he is," Arnold said.
"So ask around. Call me."
Outside, Fell said, "You're a scary sonofabitch sometimes. You sorta used your friend . . ."
"My friend's dead, she doesn't care," he said. And he shrugged. "But hillbillies understand that revenge shit."
"What's the name?"
"Lew Whitechurch. And she thinks he might deal pills."
"Let's get him," Fell said. As they were flagging the cab, she said, "If I bust Bekker myself, I'll make detective first before I get out."
"That'd be nice." A cab zigged through the traffic toward them.
"More pension. I could probably afford a straight waitress job. I wouldn't have to dance topless," Fell said.
"Aw," he said. "I was planning to come down for your first night."
"Maybe we could work something out," she said, and climbed into the cab before he could think of a comeback.
They caught Lewis Whitechurch pushing a tool cart through a basement hallway at Bellevue. His supervisor pointed him out, the hospital's assistant administrator hovering anxiously in the background. Kennett's people had been there earlier, had talked to two employees, she said, but not Whitechurch.
"What?" Whitechurch said.
Fell flashed her badge, while Lucas blocked the hall. "We need to talk to you, privately."
Whitechurch shook his head. "I don't want to talk to anyone."
"We can talk here or I can call a squad and we can go over to Midtown South."
"Talk about what?" Whitechurch shot a glance at the supervisor.
"Let's find a place," Lucas suggested.
They found a place in the hospital workshop, sitting on battered office chairs, Whitechurch spinning himself in quarter-turns with the heel of one foot. "I honest to God don't know . . . ."
Five hundred cases of paper, they said.
"I ain't gonna talk about nothing like that," he said, his Jersey accent as thick as mayonnaise. "You want to talk about this other guy, Bekker, I'd help you any way I can. But I don't know nothing about him, or any medical gear. I wouldn't touch that shit . . . ." He caught himself. "Listen, I don't take nothing out of here, but if I did, I wouldn't take that stuff. I mean, people might die because of it."
"If we catch the guy who's helping Bekker . . . that guy's going down as an accessory. Attica, and I'll tell you what, man: there'd be no fuckin' parole, not for somebody who helped this asshole . . . ."
"Jesus Christ, I'd tell you," Whitechurch said. He was sweating. "Listen, I know a couple of people who might know something about this . . . ."
"What do you think?" Fell asked.
"He covered himself pretty well. I don't know. We got names, anyway. We'll come back to him. Let him stew . . . ." Whitechurch had given them two more names. Both men were working.
"Jakes is an orderly—he oughta be around," the assistant administrator said. She was getting into the hunt, falling into Fell's laconic speech pattern. "Williams—I'll have to look him up."
They found Harvey Jakes moving sheets out of the laundry.
"I don't know about this shit," he said. He was worried. "Listen, I don't know why you'd come looking to me. I never been up on anything, never took anything, where'd you get my name . . ."
Williams was worse. Williams worked in the laundry, and was stupid. "Said what?"
"Said you boosted stuff out of here and . . ."
"Said what?"
Lucas looked at him closely, then at Fell, and shook his head. "He's not faking."
"What?" Williams looked slowly from one to the other, and they sent him back to his laundry.
"We're into a black market—pretty casual, hard to pin down, picking up the occasional opportunity," Fell said as they ambled down the hall. Like the rest of New York, the Bellevue interior was mostly a patch, painted white with black trim. "Doesn't feel like a real tight ring. Whitechurch might be bigger, if he really organized a truck to haul that paper out of there. Jakes and Williams are small-time, if they're stealing anything at all."
"That's about right," Lucas said. "Whitechurch might be something, though."
"Want to go back on him?"
"We should," he said, sticking his hands in his pockets. "But I fuckin' hurt . . . ."
"You keep poking at your cheek," she said. She reached out and touched the bruise, and her light hand didn't hurt at all. "So what are we doing?"
"I'm going back to the hotel. I need a nap, I feel like shit," Lucas said.
"We're stuck?"
"Except for Whitechurch, I don't know where we go," Lucas said. "Let's think about it. I'll call you tomorrow."
## CHAPTER
11
At the Lakota, Lucas examined his swollen cheek in the mirror. The color of the bruise was deepening, a purple blotch that dominated the side of his face, shiny in the middle, rougher toward the edges. He touched the abraded skin and winced. He'd been hit before, and knew what would happen: the abrasions would scab over while the skin around them turned yellow-green, and in a week, he'd look even worse; he'd look like Frankenstein. He shook his head at himself, tried a tentative grin, ate a half-dozen aspirin and slept for two hours. When he woke, the headache had faded, but his stomach was queasy. He gobbled four more aspirin, showered, brushed his teeth, fished an oversize Bienfang art pad from under the bed and got a wide-tipped Magic Marker from his briefcase. He wrote:
> _Bekker._
> Needs money.
> Needs drugs.
> Lives Midtown w/friend?
> Has vehicle.
> Hasn't been seen. Disguise?
> Chemist skills.
> Medical skills.
> Contact at Bellevue.
> Night.
He tacked the chart to a wall and lay on his bed, studying it. Bekker needed money if he was buying drugs, and he almost certainly would be. In the Hennepin County Jail he'd begged for them, for chemical relief.
Therefore: he had to be talking to dealers, or at least one dealer. Could he be working for one? Not likely as a salesman: even the dumbest of the dumb would recognize him as a time bomb, if they knew who he was. But if he was working as a chemist—methedrine was simple to synthesize, with the right training and access to the raw materials. If he were running a crank line, that would explain where he'd get money, and drugs, and maybe even a place to stay.
The car was another problem. He was dumping the bodies, obviously from a vehicle. How would he get access? How would he license it? Everything pointed to an accomplice . . . .
He stood, wandered into the bathroom and looked in the mirror. The abrasion was stiffening. He probed it with a fingernail, lifting a flake of skin, and blood trickled down his cheek. Damn. He knew better. He got a wad of toilet paper, held it to his cheek, and went back to the bed.
He looked at the chart again, but his mind drifted away from Bekker, toward the other case. Why had they jumped him? And had they really gone after him, or was something else happening? They could have taken him with guns: they had him cold. If they hadn't wanted to kill him, they still could have gotten to him more quickly, with baseball bats. Why had they risked resistance? If he'd had a gun in his hand, he would have killed them . . . .
Why had Lily looked out the window when she did?
But the major puzzle was more subtle. He wasn't getting anywhere, and Lily and O'Dell must see that. All he could do was look at paper, and listen to people talk. He had none of the insider information, the history, that could point him in the right direction. And yet . . . he was surrounded by people who might be involved: Fell, Kennett, O'Dell himself, even Lily. And not coincidentally.
At eight-thirty he got up; he dressed, went out to the street, flagged a cab, and rode ten minutes to Lily's apartment. She was waiting.
"You still look rough," she said as she opened the door. She touched his cheek. "Feels hot. Are you sure you want to do this? It's a lot of running around."
"Yeah," he said, nodding. "Rich is set for nine?"
"Yes. He's nervous, but he's coming."
"I don't want him to see me," Lucas said.
"Okay. You can sit in the kitchen with the lights out, talk to him down the hall."
"Fine." Lucas, hands in his pockets, wandered down toward the kitchen.
"Anything new on Bekker?" she asked, trailing behind.
"No. I was thinking, though, he must be out only at night." Lucas perched on a tall oak stool and leaned on the breakfast bar. A handicraft ceramic bowl full of apples sat on the bar, and he picked one of them up and turned it in his fingers. "Even with stage makeup, his face would be too noticeable in daylight."
"So?"
"Would it be possible to make random stops of single men driving inexpensive cars, after midnight, Midtown?"
"Jesus, Lucas. The chance of picking him up that way would be nil—and we'd probably get three cops shot by freaks in the meantime."
"I'm trying to figure out ways to press him," Lucas said. He dropped the apple back in the bowl.
"Do we really want to chase him out of here? He'd just go somewhere else, start again . . . ."
"I don't know if he can. Somehow, I don't know how, he's got a unique situation here. He can hide, somehow," Lucas said. "If he travels, he loses that—I mean, look, right now Bekker's one of the most famous people in the country. He can't go to motels or gas stations, he can't take any kind of public transportation. He can't really ride in a car without a lot of tension—if he gets pulled over by a cop, he's done. And he needs his dope, he needs his money. If we pushed him out, if he tried to run, he'd be finished."
She thought about it, then nodded. "I suppose we could do something. I wouldn't want to make a lot of stops, but we could _announce_ that we are, and ask for cooperation from the public. Maybe make a couple of stops for the TV crews . . ."
"That'd be good."
"I'll talk to Kennett tomorrow," she said. She perched on a stool opposite from him, crossed her legs and wrapped her hands around the top knee.
"How'd he get on this case? Kennett?" Lucas asked.
"O'Dell pulled some strings. Kennett's one of the best we've got on this kind of thing, organizing and running it."
"He and O'Dell don't like each other."
"No. No, they don't. I don't know why O'Dell pulled him, exactly, but I can tell you one thing: he wouldn't have done it unless he thought Kennett would find Bekker. Back in Minneapolis, you can control the bureaucratic fallout, because the department's small and everybody knows everybody else. But here . . . We've got to find Bekker, or heads'll start rolling. People are pissed off."
Lucas nodded, thought about it for a second, and said, "Kennett's an intelligence guy: are you sure he's not involved with Robin Hood?"
Lily looked down at her hands. "In my heart, I'm sure. I couldn't prove it, though. Whoever's running this thing must have a fair amount of charisma, to hold it together, and good organizational skills . . . and certain political opinions. Kennett fits."
"But . . . ?"
"He has too much sense," Lily said. "He's a believer in, what? Goodness, maybe. That's what I feel about him, anyway. We talk about things."
"Okay."
"That's not exactly proof," Lily said. She was tight, unhappy with the question, chewing at it.
"I wasn't asking for proof, I was asking for an opinion," Lucas said. "What about O'Dell? He seems to be running everything. He runs you, he runs Kennett. He's running me, or thinks he is. He picked Fell out of the hat . . . ."
"I don't know, I just don't know. Even the way he picked Fell, it seems more like magic than anything. We may be on a complete wild-goose chase." She was about to go on, but chimes sounded from the door. She hopped off her stool and walked down the hall and pushed her intercom button. A man's voice said, "Bobby Rich, Lieutenant."
"I'll buzz you in," Lily said. To Lucas, she said, "Get the lights."
Lucas turned off the lights and sat on the floor, legs crossed. Sitting in the dark, he watched Lily as she waited by the door, a tall woman, less heavy than she once was, with a long, aristocratic neck. _Charisma. Good organizational skills. Certain political opinions._
"How did you talk O'Dell into bringing me here?" he asked abruptly. "Was he reluctant? How hard did you have to press?"
"Bringing you here was more his idea than mine," Lily said. "I'd told him about you and he said you sounded perfect."
Rich knocked on the door as Lucas thought, _Really?_
Rich was a tall black man, balding, athletic, hair cropped so closely that his head looked shaven. He wore a green athletic jacket with tan sleeves, and blue jeans. He said, "Hello," and edged inside the apartment. Lily pointed him at a chair where Lucas could see his face, and then said, "There's another guy in the apartment, in the kitchen."
"What?" Rich, just settling on the chair, half rose and looked down the hall.
"Don't get up," Lily snapped. She pointed him back into the chair.
"What's going on here?" Rich asked, still peering toward the kitchen.
"We have a guy who's getting close to Robin Hood. Maybe. He doesn't want you to see his face. He doesn't know who to trust . . . . If you don't want to talk about it, with him back there, we can cut it off right here. You can go back into the bedroom while he leaves. Then it'll be just you and me . . . but I wanted you to know."
Rich's tongue slid over his lower lip, his hands gripping the arms of the chair. After a minute, he relaxed. "I don't see how he can hurt me," he said.
"He can't," Lily said. "He's mostly going to listen, maybe ask a couple of questions. Why don't you just tell me what you told Walt? If either of us has questions, we'll break in."
Rich thought about it for another moment, looked into the dark, trying to penetrate it, then nodded. "Okay," he said.
He'd been at home when he got a call from an ex-burglar he'd busted a couple of times, a man named Lowell Jackson. Jackson was trying to go straight, as a sign painter, and was doing okay.
"He said an acquaintance of his had called, a kid named Cornell, nicknamed Red. Cornell had said he'd seen Fred Waites go down and that it wasn't no gang-bangers—that one of the guys in the car was an old white guy and Cornell thought he was a cop. Jackson gave me an address."
_Old white guy?_
"Did you go after Cornell?" asked Lily.
"Yeah. Couldn't find him. So I went and talked to Jackson."
"What he say?"
"He said right after he talked to me, that same day, he saw Cornell at this playground on 118th—this is all in my report . . . ."
"Go ahead," Lily said.
"Cornell came down to a playground on 118th and said he was going home. Getting out of town. Nobody knew where he went. His last name is Reed. Cornell Reed. He's got a sheet. He's a doper, into crack. But he used to be some kind of college kid. Not a regular asshole."
"How old is he?" Lily asked.
"Middle twenties, like that."
"New York guy?"
"No. Supposedly he came from down south somewhere, Atlanta maybe. Been here a few years, though—Jackson said he didn't talk about where he came from. There was something . . . wrong. He just wouldn't talk about it. Used to cry about it, though, when he was drunk."
"How many times was he busted?"
"Half-dozen, nothing big. Theft, shoplifting, minor possession. We looked for background on him, NCIC, but there's nothing—his first busts were here in New York, addresses up in Harlem."
"And he's gone."
"Nowhere to be found. We checked Atlanta, but they don't know him."
"Dead?"
Rich frowned. "Don't think so. When he took off from the playground, he had some new shoes and a big nylon suitcase. That's what the guys at the playground say. He came up to 118th to say good-bye, they were sitting around. Then he jumped a cab and that's the last they saw of him."
"You wrote a report on all of this?"
"Yeah. And we're still looking for him. To tell you the truth, he's about the only thing we ever got on the case."
"What were you doing for Petty?" Lucas asked.
"Just looking at guys, mostly," Rich said. "Made me a little nervous, tell you the truth. I tried to get off it. I don't like looking at our own people."
"How'd you get assigned to the case?" Lucas asked.
"I don't know. Someone downtown, I guess," Rich said, his forehead wrinkling as he thought about it. "My lieutenant just said to report down to City Hall for a special assignment. He didn't know what was going on either."
"All right," said Lucas. Then, "How did Cornell know the white guy was old?"
"Don't know; if I find him, I'll ask him. Maybe just because he knows him from somewhere . . ."
They talked for another half hour, but Rich had almost nothing that wasn't in the reports. Lily thanked him and let him go.
"Waste of time," Lily said to Lucas.
"Had to try. What do you know about him? Rich?"
"Not much, really," she said.
"Good detective?"
"He's okay. Competent. Nothing spectacular."
"Hmp." Lucas touched the sore cheek, head down, considering.
"Why?"
"Just wondering," he said, looking back up. "You ready to go?"
"Want to walk? Down to the restaurant?"
"How far?" Lucas asked.
"Ten, fifteen minutes, taking it easy."
"Are we gonna get shot, going out the door?"
"No. O'Dell had a couple of people talk to the supers all along the block," Lily said. "They're looking for strange people wandering around their apartments."
The street outside the apartment was clear, but before they went out through the lobby door, Lucas scanned the windows across the street.
"Nervous?"
"No. I'm trying to figure it," he said.
She studied his face. "What?"
"Nothing." He shook his head. Rich had seemed straight enough.
"C'mon . . ."
"Nothing, really . . ."
"All right," Lily said, annoyed, still watching him.
The Village was pretty, quiet, well-tended brick townhouses with flowers in window boxes, touches of wrought iron, the image wounded here and there by a curl of concertina, a touch of razor wire. And the people looked different, Lucas thought, from the people farther uptown; a deliberate touch of the Bohemian: sandals and canvas shorts, beards and waist-length hair, old-fashioned bikes and wooden necklaces.
The Manhattan Caballero was buried in a street of red stone buildings, a small place, its name and logo painted on one window, a beer sign in the other.
"They shot from up there, the third window in, second floor," Lily said, standing on the sidewalk outside the Caballero door, pointing across the street.
"Couldn't miss with a laser sight," Lucas said, looking up at the window, then down at the sidewalk. "He must've been about right here, you see the chip marks."
Caught by the geometry and technicalities of the killing, he'd paid no attention to her. Now he looked up and she had one hand on the restaurant window, as if for support, her face pale, waxen.
"Jesus, I'm sorry . . . ."
"I'm okay," she said.
"I thought you were gonna faint."
"It's anger now," she said. "When I think about Walt, I want to kill somebody."
"That bad?"
"So bad I can't believe it. It's like I lost a kid."
They flagged a cab to go to Petty's apartment. Crossing the Brooklyn Bridge, Lily asked, "Have you ever been here? Brooklyn Heights?"
"No."
"Great place for an apartment. I thought about it, I would've come, except, you know, once you live in the Village, you don't want to leave."
"This looks okay . . ." Lucas said, peering out the window as they rolled off the end of the bridge. "The woman at Petty's apartment building . . ."
"Logan."
"She says somebody was in his apartment when he was already dead, and before the cops arrived?"
"Yes. Absolutely. She remembers that she thought he'd come home and then gone out again. She was watching television, remembered the show, and what part of the show. We checked—he'd been dead for ten minutes."
"Somebody was moving fast."
"Very fast. Had to know the minute Walt went down. Had to be _waiting_ for it. There's a question about how he got into Walt's apartment. Whoever it was must have had a key."
"That's simple enough, if you're talking about an intelligence operation."
"You should know," she said.
Petty's apartment was in a brown brick building stuck on the side of a low hill, in a cul-de-sac, the area long faded but pleasant. Marcy Logan's door was the first one to the left, inside the tiny lobby.
"Very late," Logan said, peering over the door chain at Lily's badge. She was an older woman, in her middle sixties, gray hair and matching eyes. "You said ten o'clock."
"I'm sorry, but something else came up," Lily said. "We just need to talk for a minute."
"Well, come in." Her tone was severe, but Lucas got the impression that Logan was happy for the company. "I'll have to warm up the coffee . . . ."
She had made cookies and coffee, the cookies laid out on a silver tray. She stuck a carafe of coffee in a microwave, fussed with cups and saucers.
"Such a nice apartment," Lily said.
"Thank you. They filmed _Moonstruck_ just down the way, you know. Cher was right down by the Promenade, I saw her . . ."
When the coffee was hot, Logan poked the tray of cookies in Lucas' face. Lucas tried one: oatmeal. He took another, with a cup of coffee.
"It wasn't a woman," Logan said, positively, when Lily asked. "The footsteps were too heavy. I didn't see him, but it was a man."
"You're sure?"
"I hear people come and go all day," Logan said. "That's something I'd know. I thought it was Walter coming back—I wouldn't have thought that if it was a woman."
"He went up, was there for a few minutes, then came right back down?" Lily asked.
"That's right. Couldn't have been more than a half-hour, because my show was a half-hour, and he came after the show started and left before it ended."
"You told the investigators that it occurred to you that it wasn't Petty," Lily said. "But not seriously enough that you actually looked. Why did you think it might not be him?"
"Whoever it was, stopped in the lobby. Like he was looking at my apartment door or maybe listening for anybody inside. Then he went up. Walter was always very forthright. Walked right in, went right up. Especially on his Fridays. He'd always have two or three beers, and he couldn't hold it at all, and by the time he got here, he'd . . . you know: he had to go. You could hear the water running from the toilet, right after he went up. That night, though, whoever it was stopped inside. He did the same thing on the way back out. Stopped in the lobby. It gives me the shivers. Maybe he was thinking about rubbing out witnesses."
"I don't think that's much of a threat," Lily said, smiling at the "rubbing out."
"Why don't you say something, young man?" Logan asked Lucas, who was eating his sixth cookie. He couldn't seem to stop.
"Too busy eating cookies," he said. "These things are great. You could make a fortune selling them."
"Oh, that's nice," she said, smiling. "What happened to your face?"
"I was mugged."
"Isn't that just like New York? Even the cops . . ."
"How do you know this guy went to Petty's apartment?" Lucas asked.
"Well, I heard him come in, and then the elevator dinged, so he was going up. Then just a second later, I heard another ding, like it was coming from the kitchen. That's the second floor. If it goes to the third floor, I can barely hear it. If it goes to the fourth, I can't hear it at all."
"Okay," Lucas said, nodding. "So you heard it ding on the second floor."
"Yes. And the Lynns and Golds were already in and the Schumachers were at Fire Island that whole weekend. So it had to be Walter, and it was about the time he always came in. I didn't hear him flush, though. Then I heard the elevator ding on the second floor again, and it came down. Then whoever it was, I thought was looking at apartments again, because it was a minute before the outside door opened . . . . I should have looked, but I was watching my show."
"That's fine," said Lucas, nodding. "And it wasn't a visitor to one of the other apartments?"
"No," Logan said, shaking her head. "When the cops got here and I found out what happened, I told them about somebody coming, and they talked to everybody up there. Nobody came in at that time, and nobody had any visitors."
When they finished with Logan, they rode up in the elevator and Lily cut the seals off Petty's door. The apartment had been neatly kept but had been pulled apart by investigators. The refrigerator had been unplugged, and the door stood open. Cupboard doors were open and paper was stacked everywhere. Lucas went to Petty's desk, which was set in a tiny alcove, and thumbed through financial records . . . . No personal phone book.
"No phone book."
"The Homicide guys probably have it. I'll ask."
Ten minutes later, Lily said, "This is like the interview with Rich. There isn't anything here."
On the way out, Mrs. Logan met them in the hallway with a brown paper bag, which she handed to Lucas. "More cookies," she said.
"Thanks," he said, and then, "When I finish them, I may come back for more."
The old lady giggled, and Lucas and Lily went looking for a cab.
• • •
Cornell Reed. Cornell Reed had seen the killer, an old white guy, and recognized him as a cop.
Lucas lay on the hotel bed and thought about it, sighed, rolled off the bed, found his pocket address book, and picked out Harmon Anderson's home phone number. As he dialed the number, he glanced at his watch. It would be midnight, Minneapolis time.
Anderson was in bed.
"Jesus, Lucas, what's going on?"
"I'm in New York . . . ."
"I know, I heard. I wish I was there . . . ." Lucas heard him turn away from the phone and say to someone in the background, "Lucas." Then to Lucas, he said, "My wife's here, she says hello."
"Look, I'm sorry I woke you up . . ."
"No, no . . ."
"And I don't want to cause you any problems, but would you be available to do a little computer work? I'd pay you a consultant's fee."
"Ah, fuck that, what do you need?"
"I'm in a snakepit, man. Could you find out what airlines fly out of New York, all the big airports, including Newark, and check from the beginning of the month, see if there's a ticket for a Cornell Reed? Or any first name Cornell, if you can do that? Or Red Reed? I don't think it'd be overseas, except maybe the Caribbean. Check domestic first, like Atlanta, L.A. or Chicago. I need to know where he went and I need to know who paid for the ticket, if we can find that out."
"Could take a couple of days."
"Get back to me—and I'm serious about a fee, man. A few bucks."
"We can work that out . . . ."
"Get back to me, man."
When he hung up, Lucas dropped back on the bed, thinking back to the interview with Rich. Rich didn't know why he'd been picked for Petty's team. Neither did Lily. His only qualification seemed to be that he'd later get a call from a burglar he knew, producing the only lead in the case. Good luck of a rare and peculiar variety.
Rich said that Cornell Reed was heavy into the crack. If that was right, Reed shouldn't be flying out of town. If he had enough cash to fly, he'd buy dope with the cash and take the bus. Or hitchhike. Or just not go. With enough crack, you didn't have to go anywhere . . . . He certainly wouldn't take several hundred dollars out to La Guardia and push it across the ticket counter.
On the other hand, a doper doesn't take a cab to the bus depot, not when the A train would have him there quicker and leave him enough change for a rock or two. La Guardia was another story. There was no easy way to get there, except by cab . . . .
So maybe he was flying. And maybe he was flying on an unrefundable ticket. And that sounded like a ticket issued by a government.
Or a police department.
And then there was Mrs. Logan's story.
That was very interesting; interesting and disturbing. Had Lily not understood it? Or had she hoped that Lucas hadn't?
## CHAPTER
12
Thirty hits of speed, two days; Bekker hadn't slept forever. He was carried along on the chemicals like a leaf in a river, the flow of time and thought rolling about him. And he was avoiding the woman with the eyes, the woman watching him. She terrified him: but the chemicals had defeated her after two days, and she was losing her grip.
But other things were happening.
Late in the afternoon of the second day, the bugs came. He could feel them, lines of them, inching through his veins. All of his veins, but in particular, a vein on the forearm; he could feel them, little bumps, rattling along, doing their filthy work. Eating him.
Eating the blood cells. He could remember, as a child, kicking open ant nests and seeing the ants running for cover, their mealy white eggs in their jaws. And this was the image that came to him: ants running, but with blood cells caught in their pincers. Thousands of them, running through his veins. If he could let them out . . .
A voice in his head: _No no no, hallucination, no no no_ . . .
He stood up, his knees and feet aching. He'd walked for miles in the basement, back and forth, back and forth. How far? A few errant brain cells wandered away and did the calculation . . . say five thousand round trips, twenty feet each way . . . thirty-seven point eight seven eight miles. Thirty-seven point eight seven eight seven eight seven eight seven eight seven . . .
He was snared in the eight-seven loop, captivated by the sheer infinity of it, a loop that would last longer than the sun, would last longer than the universe, would go on for . . . what?
He shook himself out of the loop, felt the bugs raging through his veins, took his forearm to the bathroom, turned on the light, looked for bumps, where the bugs scuttled along . . . .
A voice: _formication_ . . .
He pushed it away. Had to let them out, squeeze them out somehow. He walked into the operating theater, went to the instruments pan, found a scalpel, let them out . . . .
He began to walk, the bugs draining away, began to pace again—what was that smell? So clean and coppery, like the sea. Blood?
He looked down at himself. Blood was running from his arm. Not heavily, now slowing, but his hand and forearm looked as though they had been flayed. Where he'd been pacing, blood splattered the floor, an oval line marking his pattern, as though someone had been swinging a decapitated chicken.
The voice: _stereotypy._
What? He stared at the arm and a bug zipped down the vein. Like Charlie Victor on the Ho Chi Minh Trail, like Charlie Victor at the Hotel Oscar, Charlie Hotel India Mike November Lima Tango Romeo . . .
Another loop—where had that come from? 'Nam? He shook himself out. The bugs were waiting, in all their ranks.
Medication. He went to his medicine table, found a half-dozen pills. That was all. He popped one, then another. And a third.
He picked up the phone, struggled with himself, put it back down. No telephone from here, not to a dealer. Cops bugged dealers, bugged . . . He looked down at his arm, at the sticky blood . . . .
Calmed himself. Washed. Dressed. Put a bandage on the cut on his forearm. Cut? How did that . . .
He lost the thought and fixed himself in the mirror, preparing for his public, the _need_ always there, looking over his shoulder. The _need_ brought up the street personality. Changed his voice. Changed his manner. When he finished dressing, he went out to the corner, to a pay phone.
"Yes?" Woman's voice.
"May I speak to Dr. West?"
Whitechurch was there a second later. "Jesus Christ, we gotta talk. Like now. The cops have been here and they're looking for your buddy—or whoever you sold that shit to, the monitoring gear."
"What?"
"The guy you sold it to," Whitechurch said insistently. "He's this fruitcake killer, Bekker. Jesus Christ, the cops were all over me."
"New York cops?"
"Yeah, some cooze and this mean-looking asshole from Minneapolis."
"Are they on your phone?"
"This is not my phone. Don't worry about that. Just worry about the dude who bought that shit . . . ."
"I can handle that," Bekker squeaked. The effort hurt. "But I need some product."
"Jesus Christ . . ."
"A lot of it."
"How much?"
"How much do you have?"
There was a moment's silence, then Whitechurch said, "You're not with this Bekker dude, are you?"
"It wasn't Bekker. I sold it to a high school kid out on Staten Island. He's using it for his science project."
That clicked with Whitechurch: Schoolteacher . . .
Whitechurch had decided to take a vacation to Miami, could use the extra cash. "I could get you two hundred of the crosses, thirty of the angels and ten of the white, if you can handle it."
"I can handle it."
"Twenty minutes?"
"No . . . I've got to come over . . . ." Let him think Bekker lived on Staten Island. "I need a couple of hours."
"Two hours? All right. Two hours. See you at nine. Usual place."
Bekker left the Volkswagen in a staff parking ramp off First Avenue; the ramp was open to the public from six until midnight. He nodded to the guard in the booth and rolled all the way up to the top floor. He'd watched Whitechurch before. He believed in taking care and knew that drug dealers routinely sold friends and customers to the cops. He'd learned a lot in jail; another side of life.
Whitechurch insisted on punctuality. "That way, I only have the stuff on me, on the street, for a minute. Safer that way, you know?"
Usually, Whitechurch would be walking out of the hospital, or down the sidewalk toward a bus bench, when Bekker came by. Once Bekker, arriving early, had watched him from the ramp. Whitechurch had come out, walked down the sidewalk toward the bus bench, had waited for two or three minutes, then had gone back inside, using the same door he'd used on the way out. Bekker had called to apologize, and made the pickup a few minutes later.
Bekker walked down to the first floor, past the pay booth, and down the street to an alleylike passage to the emergency room. Night was settling in, the streetlights coming on. He was early, slowed down. Several people around. Not good. He turned down the alley to the emergency room, walked up to the door that Whitechurch usually came through. Pulled on the handle. Locked. Glanced at his watch. Still two minutes early. Whitechurch should be coming, just any moment . . . .
He'd done an angel before he came, part of his emergency stash. Strong stuff; it freed his power . . . .
The derringer was in his hand.
The door opened and Whitechurch stepped out, and jumped, startled, when he saw Bekker.
"What . . ."
"We've got to talk," Bekker whispered. "There's more to this than I thought . . . ."
He looked past Whitechurch to an empty tile-walled corridor. "Back inside, just a few minutes. I feel obligated to tell you about this."
Whitechurch nodded and turned, leading the way. "Did you bring the cash?"
"Yes." He held out the cash envelope and Whitechurch took it. "Have you got the product?"
"Yeah." Whitechurch turned as the metal fire door closed behind them. The corridor lights weren't strong, but they were unforgiving blue fluorescents.
Whitechurch had a plastic baggie in his hand and half stepped toward Bekker when he said, "You're . . ." He stopped, catching his tongue, and began to back away.
"The fruitcake killer," Bekker said, smiling. "Just like on _I've Got a Secret._ You remember that show? Garry Moore, I think."
Whitechurch's head snapped around, looking for room, then turned back to Bekker, but already his body was moving, trying to run.
"Listen," he said, half over his shoulder.
"No." Bekker leveled the gun at Whitechurch's broad back and Whitechurch shouted, "No way," and Bekker shot him in the spine between the shoulder blades. The muzzle blast was deafening, and Whitechurch pitched forward, tried to catch himself on the slick tile walls, bounced and turned. Bekker pointed the pistol at him, from two feet.
"No way . . ."
Bekker pulled the trigger again, firing into Whitechurch's forehead. Then he pushed the gun into his pocket, hurrying, took out a scalpel, stooped, and ruined Whitechurch's dead eyes. Good.
Down the hall, a door banged open. "Hey." Somebody yelling.
Bekker looked down the corridor: empty. He grabbed the baggie full of pills, stood, remembered the money, saw it half trapped under Whitechurch. Down the hall, the door banged open again and Bekker jerked at the money envelope. The envelope ripped, but he got most of it, just a bill or two still trapped under the body.
"Hey . . ." He looked back as he went through the door, but there was nothing in the corridor but the voice. Outside, he gathered himself and hurried, but didn't run, down the alley, turning left on the sidewalk to the parking ramp. He went inside to the stairs, heard footsteps behind, and half turned.
A young woman was hurrying after him. He started up the steps and she caught up with him, a few steps behind. "Wait up . . ." Breathless. "I hate to go up here alone. If there were somebody . . . You know."
"Yes." The woman was worried about being attacked. There was only one open entrance to the ramp, but anyone could get in over the low walls. Judging from the graffiti spray-painted on the concrete walls, several people had.
"God, what a day," the woman said. "I hate to work when it's so nice outside, I never see anything but computer terminals."
Bekker nodded again, not trusting his voice. If he'd had the time, he could have taken her. She'd have been perfect: young, apparently intelligent. A natural observer. Might possibly understand the privilege she was being given. He could take her, he thought. Right now. Hit her in the head . . .
Behind her, he balled his hand into a fist, and he thought, _Or the gun. I could use the gun._ He felt the weight of the gun in his pocket. Empty now, but a threat . . .
But if he hurt her, struck her, had to fight, if she was less than a perfect specimen . . . his results would be impeachable. People were watching him, people who hated him, who would do anything to impeach his results. He fell back a step, his heart beating like a drum.
"See you," she said, one half-level below his car. She looked out on the open floor before she went through the door. "Nobody here . . . makes you feel a little stupid, doesn't it?"
He could, but . . . wait. No improvisation. _Remember the last time . . . Easy, easy, there are plenty of them._
Bekker lifted a hand and risked it: "Good-bye," he said, in his careful voice.
He had to get one. Had to. He didn't realize, until he saw the woman get in the car and lock the doors, how strong the need was now.
He rolled out of the ramp, straight down the street; there was some commotion in the emergency entrance alley, but he didn't stop to look. Instead, he went straight back to his apartment, almost frantic now, and got out his collector's bag: the stun gun and the anesthetic tank and mask. He flicked the stun gun once, checked the discharge level. Fine. And dug through the bag he'd taken from Whitechurch: just a taste. He snapped one of the angels between his teeth, thinking to take a half, but a half wouldn't do, and he took a whole, waiting for the power to come.
Cruising, thinking: _Infrared. Ultraviolet. Breakthrough._
He knew this bar . . . .
Later. He saw the woman slouch out of the back of the bar, lean against the brick exterior, and light a cigarette with what looked like an old-fashioned Zippo. Not many men around, lots of women coming and going, many of them alone. Easy targets.
The woman was leaning against the outside wall, wearing jeans and a sleeveless T-shirt, with a wide leather belt. She had short black hair, with gold hoop earrings.
Bekker came up, stepping carefully around the Volkswagen as though he didn't own it. Not too aggressive. Stun gun in his hand, tank under his arm, hand on mask.
"Terrific night," he said to the woman.
She smiled. "You're looking pretty good," she said.
Bekker smiled back and stopped next to the nose of the Volkswagen.
_Come to the gingerbread house, little girl_ . . .
## CHAPTER
13
"What's wrong?" Lily asked.
Kennett rolled toward her and put an arm under her head. "I feel like an invalid when we do that. I mean, nothing _but_ that."
The forward double berth was wedge-shaped, shoved into the bow of the boat. Kennett was lying on his side. He reached toward her face in the near-darkness, touched her at the hairline with the pad of his index finger, drew it down her nose, gently over her lips, between her breasts, then up to gently tap each nipple, then down around her navel, over her hipbone and down the inside of her thigh to her knee. She was still warm, sweating.
"We're not . . . compelled . . . to do it," Lily said.
"Maybe you're not, but I am," Kennett grumbled. "If I couldn't make love anymore, I'd feel like a goddamn vegetable."
"You just wanna be on top," she said, trying to make a joke out of it. When he didn't respond, she said, "You've got to listen to Fermut."
"Fuckin' doctors . . ." Fermut, the cardiologist, had reluctantly agreed that Kennett could resume his sex life "as long as your partner does the hard work."
"Listen to him," Lily said, gently but urgently. "He's trying to save your life, you dope."
"Yeah." Kennett turned his head away from her, his hand scratching at his chest.
"You want a cigarette, right?"
"No, that's not it. I was just thinking . . . it's not the doctors. It's me. When I get turned on and my heart starts thumping, I start listening to it . . . ."
"Then we oughta quit. Maybe only for a few weeks . . ." Lily said.
"No. That'd be worse. It's just . . . Christ, I wish one thing—just one goddamn thing in this world—was simple. Just one thing. I gotta get laid, but if I get laid, I can't help thinking about my heart, and that can mess up getting laid. Then with you on top all the time, and me just laying there like a dead man with a hard-on, I start thinking, what's it like for her? It must be like necrophilia, screwing me."
"Richard, you idiot . . ."
"Christ, I'm glad I met you," he said after a while. "I couldn't believe you were in there, working for O'Dell. I kept thinking, she can't be just working for him, a woman like that, there's gotta be something else going on here."
"Oh, God . . ." Lily giggled, an odd, pleasant sound with her husky voice.
"Sorry 'bout that," Kennett said, touching her again. "I wonder what O'Dell does for sex? Fly out to Vegas and get a couple-three fat ones in the sack? I wonder how long it's been since he's seen his dick? He's so fat I don't think he can even reach it anymore . . . ."
"C'mon . . ." Lily said, but she giggled again, a big woman giggling, and that set Kennett off, laughing.
And then: " 'Course, things must've been different with Davenport."
Lily cut him off: "Shut up. I don't want to hear it."
"Probably hung like a Shetland pony . . ."
"You wanna get bit?"
"Is that a clear offer?"
"Dick . . ."
"Hey. I'm not jealous. Well, maybe a little. But I really like the guy. This whole business of bringing him to dance with the media, that's pretty bizarre, and it's working. You think he'll get in the sack with Barbara Fell?"
"I don't know," she said, crisply.
"He seems like the kind of guy who'd be looking around," he offered.
"Pot and kettle."
"Hey—I didn't say it was bad. I just wondered about him and Fell. That's a match made in hell."
"She's very attractive."
"I guess, if you like the type," Kennett said. "She looks like a biker chick who fell off the Harley one too many times. Why'd you put him with her? Some kind of psychological compulsion to bury your sexual history?"
"No, no, no. We just needed somebody who knew Midtown fences . . . ."
"Yeah, but Davenport's supposed to be a talking head."
"He's never a talking head. Even when he's talking. The guy has more moves than you do, and you're the sneakiest, shiftiest . . ."
" . . . crookedest . . ."
" . . . most underhanded asshole on the force. Besides, he had to do something to get the media to talk to him."
"I suppose." Kennett's fingertips slipped along her thigh again, her skin soft and slightly cool from evaporating sweat. "We'll either have to get a sheet to cover up or figure out some way to warm up the place again."
Lily groped for his groin and said, "Oh, Jesus. Are you sure? Dick . . ."
He rolled into her, his arm around her, pulling her tight. "That's the word, all right. Dick."
"Be serious."
"All right. How's this: I really do need you; it's the thing that keeps my heart going . . . ."
Much later, when he was sleeping, she thought: _They can all make you feel guilty; it's what they do best._ . . .
## CHAPTER
14
The phone rang early and Lucas rolled out of the blankets, dropped his feet to the floor and sat a moment before he picked up the receiver. "Yeah?"
"How's your head?" Kennett sounded wide awake and almost chipper.
"Better," Lucas said. He couldn't seem to focus and noticed that the window shade was bright with low-angle sunshine. "What time is it?"
"Seven o'clock."
"Ah, Jesus, man, I don't get up at seven . . . ." His face hurt again, and when he turned toward the bed, he noticed a spot of blood on the pillowcase.
"Hey, it's a great day, but it's gonna be hot," Kennett said cheerfully.
"Thanks. If you hadn't called, I woulda had to look out the window . . . ." _What's going on?_
"I understand that you and Fell talked to a guy named Whitechurch yesterday, at Bellevue?"
"Yeah?"
"Bekker took him off last night."
"What?" Lucas stood up, trying to understand.
"Shot him in a hallway. Cut his eyes," Kennett said.
"The morgue guys said it's gotta be Bekker, 'cause it was done too well to be a copycat. And with you talking to him about Bekker, there's no way it's a coincidence. When they called me, a couple of hours ago now, I shipped Carter over to the hospital. Somebody there finally figured out that cops were talking to Whitechurch yesterday . . . ."
"Ah, Jesus," Lucas said. "Whitechurch was wrong, too. We knew it. We knew he was bullshitting us."
"How'd you get onto him?"
"A fence," Lucas said. "Down on the Lower East Side."
"Smith?"
"No, a small-timer, a woman named Arnold. We'll go back and talk to her, but I don't think she has any connection with Whitechurch except to handle occasional shipments from him. But why was Bekker talking to Whitechurch again? More equipment?"
"Whitechurch was dealing dope," Kennett said.
"Ah. For sure?"
"Yeah, we got it from a couple of places. And I'd bet that's where the halothane is from."
"Telephones?"
"We sent a subpoena over, and the phone company's mopping up their computers right now. They'll run back all the calls that came into Whitechurch's apartment and his office phone, both, and where they came from, for the last two months."
"That should do it," Lucas said. "Fell's got a beeper: if you find him, call us. I'd like to see the end of it."
"Mmm. It doesn't feel that easy," Kennett said.
"All right. Well: I'll get Fell and get back to the fence. Goddammit, why'd Whitechurch cover for him? That'd be something to figure out."
Lucas called Fell and told her.
"Did we mess it up?" she asked anxiously.
"No. We barely touched the guy—there was no way to know. But Kennett's people are all over him now. Everybody who knew him. We've got to talk to what's-her-name, the fence."
"Arnold. Rose."
"Yeah . . . So what's your status? Are you ready?" Lucas asked.
"Hey, I'm just sitting here on my bed, buck naked, half asleep."
"Jesus, if you had a warm croissant and a cup of coffee, I'd come right over," Lucas said. The nude photo of Fell and the other cop popped up in his head.
"Fuck you, Davenport," Fell said, laughing. "If you're ready, why don't you get a cab? I'll be out front by the time you get here."
"You come get me," Lucas said. "I'm barely awake, and I gotta shave." He touched his raw cheek.
"Be ready," she said.
Fell, when she arrived, was wearing a black tailored cotton dress with small flowers—the kind of dress women wore in Moline, Illinois—black low heels and nylons.
"Jesus, you look terrific," Lucas said, climbing into the cab behind her.
She blushed and said, "We just gonna walk in on Arnold?"
"You don't want to talk about how terrific you look?"
"Hey, just shut the fuck up, okay, Davenport?" she said.
"Anything you want . . ." Under his breath, he added, "Toots."
"What? What'd you just say?"
"Nothing," Lucas said innocently.
She closed one eye and said, "You're walking on the edge, buddy."
Arnold was scared. "He maybe got done because he talked to you," she said, sucking her heavy lips in and out.
"No. He got done because he called this asshole Bekker, who he was protecting, and told him that we'd interviewed him," Lucas said. "Bekker knows me. He didn't want to take any chances."
"So what do you want from me? I gave you everything."
"How'd you get in touch with Whitechurch when you needed to?" Lucas asked.
"I never needed to. When he had something good, he'd bring it over. Otherwise—shit, I don't handle hospital stuff. I handle shit you can sell, cheap. Suits. Neckties. Telephones. I wouldn't know what to do with no hospital stuff."
Fell pointed a finger at her: "You took down Simpson-McCall, what, two months ago . . . ?"
Arnold looked away. "No. I don't know nothing about that."
Fell studied her for a moment, then looked at Lucas. "Brokerage moves to a new building, one of those over-the-weekend moves. Trucks coming and going all night with files, computers, telephones, furniture, putting it in. The only thing is, not all of the trucks were hired by the brokerage. Some assholes rented trucks, drove them up to the loading docks, and disappeared over the horizon. . . . One of them took off six hundred brand-new beige two-button phones. Somebody else got fifty Northgate IBM compatibles, still in the boxes."
"Really?" said Arnold, faintly distressed. "Computers?"
Fell nodded, and Lucas looked back at Arnold. "If you _had_ to get to Whitechurch, what'd you do?"
Arnold shrugged. "Call him at the hospital. Wasn't no big secret where he worked. Nights only, though."
"Did he have a special number?"
"I don't know, man, I never called him."
"Did . . ."
Fell's beeper went off. She took it out of her purse, glanced at the readout. "Where's the phone?" she asked Arnold. To Lucas, she said, "I bet they got him."
"Over there, at the end of the counter, underneath . . ." Arnold said, pointing.
As Fell punched the number into the telephone, Lucas went back to Arnold. "Did he work with anybody?"
"Man, I bought telephones from him, four dollars apiece," Arnold said impatiently. "Boxes of pens and pencils. Notepads. Cartons of Xerox paper. Cleaning supplies. He once came in with two hundred bottles of ERA, you know, the laundry soap. I don't know where he got it, I didn't ask any questions. And that's all I know about him."
"Yeah, this is Fell, you beeped?" Fell said into the phone. And then, voice hushed, "Jesus. What's the address. Huh? Okay." She hung up and looked at Lucas. "Bekker did another one, another woman. Ten minutes from here, walking."
Lucas pointed a finger at Arnold: "Did you hear that? Think about Whitechurch. Anything you think of, call us. Anything."
"Man, there's nothin' . . ."
But Lucas and Fell were out the door.
The body was in a dead-end alley off Prince. Uniforms blocked the mouth of the alley, kept back the curious. Fell and Lucas flashed their badges and went through. Kennett and two other plainclothesmen were there, staring into a window well. Kennett's hands, gripping the rail around the well, were white with tension.
"Goddamn maniac," he said as Lucas and Fell walked up. The crime-scene techs had dropped a ladder into the well. Lucas looked over the railing and saw a small woman's body at the bottom of the well, nude, crumpled like a doll, the techs working over her.
"No question it was Bekker?" Lucas asked.
"No, but it's different. This doesn't look so scientific. She's pretty slashed up, like he . . . I don't know. It looks like he was having fun."
"Eyes?"
"Yeah, the eyes are cut and the doc says it looks like his work. The eyelids gone, very neat and surgical. The sonofabitch has a signature."
"How long has she been down there?" Fell asked.
"Not long. A few hours at the most. Probably went in before dawn, this morning."
"Got an ID?" asked Lucas.
"No." Kennett looked at Fell, who was lighting a Lucky. "Could I bum one, I . . ."
"No." Fell shook her head, carefully not looking at him.
"God damn it," Kennett said. He stuck one hand in his jacket pocket, put two fingers of the other between his shirt buttons, over his heart. He caught himself, pulled them out, looked at his hand and finally stuck it in the other jacket pocket. "Fuckin' do-gooders."
"Anything on the Bellevue phones?" Lucas asked, watching the techs get ready to roll the woman's body.
Kennett's forehead wrinkled. "Think about this, Davenport: We got a guy who deals drugs, but he gets no phone calls. I mean, like, almost none. He got six calls at his apartment last month. There was a phone in the maintenance office he could use, but he didn't, much. At least, that's what his supervisor says."
"Did he carry a beeper? Maybe a cellular?" Fell asked.
"Not that we can find," said Kennett.
"That's bullshit," Lucas said flatly. "He was dealing, right? We know that for sure?"
"Yeah."
"Then he's got a phone. We've just got to find it . . . ."
"Carter's guys are interviewing people over there right now, at Bellevue. Maybe you could listen in for a while?" Kennett said. He looked at Fell. "You're the only guys who've come up with anything."
At the bottom of the window well, the crime-scene techs rolled the body. The woman's head flopped over, and her wide white eyes suddenly looked up at them.
"Aw, shit," Fell gagged. She turned away, hunched over the alley cobblestones, and a stream of saliva poured from her mouth.
"You okay?" Lucas asked, his hand on her back.
"Yes," she said, straightening. "Sorry. That just caught me, the eyes . . ."
Five minutes later, the body was out of the window well. The removal crew had wrapped it in a blanket, but Kennett ordered the wrapper peeled away. "I want to look," he said evenly. "I wish the fuck I could have gotten down there . . . ."
Kennett and Lucas squatted next to the collapsible gurney as the blanket was lifted. The woman's face was like marble, white, solid, her dying pain and fear still graven on her face. The gag was like the earlier ones, carved from hard rubber, held in place with a wire that had been twisted tight behind her ear.
"Pliers," Kennett said absently.
"Treats them like . . . lumber," Lucas said, groping for the right concept.
"Or lab animals," Kennett said.
"Sonofabitch." Lucas leaned to one side, almost toppled, caught himself with his hand, then knelt over the body until his face was only inches from the body's left ear. He looked up at one of the techs and said, "Roll her a little to the right, will you?" He took a pen from his shirt pocket and, to Kennett, said, "Look at this."
Kennett knelt beside him and Fell squatted behind the two of them, the other detectives crowding in. Lucas used the pen to point at two oval marks on the dead woman's neck muscle.
"Have you ever seen anything like that?" Lucas asked.
Kennett shook his head. "Looks like a burn," he said. "Looks like a fuckin' snakebite."
"Not exactly. It looks like a discharge wound from one of those electroshock self-defense gizmos, stun guns. The St. Paul cops carry them. I went over to see a demonstration. If you keep the discharge points on bare skin for more than a second or two, you can get this kind of injury."
"That's why there's no fight," Fell said, looking at him.
Lucas nodded. "He hits them with the shocker. When you get hit, you go down, like right now. Then he comes with the gas."
"Couldn't be too many places around that sell those things," Kennett said.
"Police-supply places, but I've seen them in gun magazines, too, mail order," Lucas said.
Kennett stood and rubbed alley sand from his hands and tipped his head back, as though looking up to heaven. "Please, God, let me find a Midtown address on an order form."
Lucas and Fell took a cab to Bellevue, windows open, the hot popcorn smell of the city roaring in as they dodged through traffic, and got trapped for five minutes in a narrow one-lane crosstown street. Fell's jaw was working with anger.
"Thinking about Bekker?"
"About the body . . . Jesus. I hope Robin Hood gets him," she said. "Bekker."
"What? Robin Hood?" He looked at her curiously.
"Nothing," she said, looking away.
"No, c'mon, who's Robin Hood?"
"Ah, it's bullshit," she said, digging in her purse for a cigarette. "Supposedly somebody is knocking off assholes."
"You mean, a vigilante?"
She grinned. "How else you gonna run this place?" she asked, gesturing out the window. "It's supposed to be cops, but I think it's just bullshit. Wishful thinking."
"Huh."
She lit the cigarette, coughed, and looked out the window.
Whitechurch had been a maintenance foreman. A changing roll of a dozen people worked under his loose supervision, doing minor repairs all over the hospital on the three-to-eleven shift.
"A great goddamn job if you're stealing stuff," Fell said as they joined Carter in an employees' lounge. Three detectives were interviewing hospital employees, with Carter supervising.
"Or if you're dealing," said Carter. He looked at his list. "Next one is Jimmy Beale. Goddamn, I got little faith in this."
"I know what you mean," Lucas said, watching the scared employees trooping through the lounge.
Beale knew nothing. Neither did any of the rest. Fell burned through a pack of Luckys, left to get another, came back and leaned in the door.
"God damn it, Mark . . . it's Mark?" Carter was saying. "God damn it, Mark, we're not getting anywhere and it's hard to believe that a guy could be stealing the place blind and nobody'd know about it. Or dealing dope, and nobody'd know . . . ."
Mark, tall, narrow, acned, nodded nervously, his Adam's apple working convulsively, sliding up and down his thin neck. "Man, you never seen the dude, you know? I mean, I'd come in and he'd say, Mark, g'wan up to 441D and put on a new doorknob and then see if there's a leak on the drinking fountain up on six, and that's what I'd do. He'd come by, but like, I never hung out with him or nothing."
When he was gone, Lucas said, "Nobody knew. How many do you believe?"
"Most of them," Carter said. "I don't think he was dealing here. And if you're stealing stuff, you don't talk about it. Somebody'll try to cut in—or somebody'll try to do the same thing, then feed you to the cops on plea bargain."
"Somebody must've known," Fell objected. "That was the last of them?"
"That was the last . . ." said Carter.
A woman knocked on the edge of the door and stuck her face in. She had curly white hair and held her hands in front of her as though she were knitting.
"Are you the police?" she asked timorously.
"Yeah. C'mon in," Lucas said. He yawned and stretched. "What can we do for you?"
She stepped inside the room and looked nervously around. "Some of the others were saying you were asking if Lew had a beeper or a walkie-talkie?"
"Yes. Who are you?"
"My name is Dotty, um, Bedrick, I work in housekeeping?" She made her sentences into questions. "Last week, Lew split out his pants, right down by housekeeping? There was some kind of pipe thing he was working on and he bent over and they went, split, right up the back?"
"Uh-huh," Lucas said.
"Anyway, I was right there? And everybody knows I sew, so he came in and asked if I could do anything? He slipped right out of his pants—he was wearing boxer shorts, of course—he slipped right out and I sewed them up. He was just wearing a T-shirt on top, and the boxer shorts, and I had his pants. There was nothing in there but his wallet and his keys and his pocket change. There wasn't any beeper or anything like that."
"Hey. Thank you," Lucas said, nodding. "That was a problem for us."
"Why did you have to know?" Bedrick asked. Lucas thought, _Miss Marple._
"We think that—I'm sure you've heard this from the others—we think he was dealing drugs. If he was, he needed access to a telephone."
"Well, there was something odd about the man . . . ."
She wanted to be led: Lucas put his hands on his waist, pushing his sport coat back on both sides, like a cop on television, let a hip pop out and said, "Yeah?"
She approved: "Sometimes when the calls came over the speakers for doctors, I've seen him look up at the speakers. And the next thing, _he'd_ be calling in. I saw him do it two or three times. Like he was a _doctor._ "
"Sonofagun," Carter said. "There'd be a call for a doctor?"
"That's right."
"Jesus," he said, turning to Lucas and Fell, dumbfounded. "That's it."
"That's it?" chirped Bedrick.
"That's it," Carter said. He smiled at the old lady and shook his head. "I never had a civilian do that before."
Fell decided to stay at Bellevue and work the lead. Lucas, shaking his head, decided to head back to Midtown South.
"You don't think it'll be anything?" Fell asked.
"It might be—but with Whitechurch dead, I don't know how you'd find out," he said.
"I want to stay anyway," she said. "It's all we've got."
_All we've got,_ Lucas thought. _Yeah. We find Bekker's supplier, the best damned lead all week, and Bekker kills him right under our noses._ Some hotshot cops they were. There had to be another way to approach this situation, to find a way in . . . .
At Midtown South, Lucas could hear Kennett all the way out to the reception desk.
" . . . know it's hot, but I don't give a shit," he was saying. "I don't want people around here reading the goddamn reports, I want everybody out on the street. I want the fuckin' junkies to know there's a war going on. Instead of coming in here, I want you out on the street with your people, rousting these assholes. Somebody knows where he's at . . . ."
Lucas leaned in the door. Seven or eight detectives were sitting uncomfortably around the conference room, while Kennett sat on a folding chair at the front, his fingers over his heart, an angry flush on his face. He looked over the cops to Lucas and snapped, "Tell me something good."
"Did you talk to Carter?"
"I'm supposed to call him back," Kennett said, looking at a phone slip. "What happened?"
"An old lady maybe told us how Whitechurch got his calls."
"Well, goddamn," somebody said.
Lucas shook his head: "But it might not be good. He may have had doctor code names for his clients. When a buyer needed to call in, the switchboard—or somebody—would page the doctor. Whitechurch would pick up a phone and answer the page. There are thousands of doctors in there every day, thousands of phone calls. Hundreds of pages."
"Sonofabitch," Kennett said. He ran his hand through his hair, and a swatch of it stood up straight, in a peak. "Carter's pushing it?"
"Yeah. Six guys and Fell stayed to help."
Kennett thought about it for a second, then exhaled in exasperation and asked, "Anything else?"
"No. I'm still reading paper on him, but I think . . . Look, I had an idea on the way over. Entirely different direction. Carter's taking the phone angle, you got guys on everything else. I was thinking again about how hard Bekker is to find, about where he's getting his money, about all the things we don't know about him. So I was thinking, maybe I should talk to the guys who _did_ know Bekker."
"Like who?"
"Like the guys who were in jail with him. Maybe I ought to go back to the Cities. I could run down the people who were in the next cells to his. Maybe he said something to somebody, or somebody gave him an idea of how to hole up . . . ."
"That's not bad," said Kennett, scratching his breastbone. "Kind of a long shot, though, and it takes you out of the action here." He thought about it some more. "I'll tell you what. Read paper for the rest of the day, think about the phones. Day after tomorrow's the lecture. If we've got nothing by then, let's talk about it . . . . You see the art?"
"Art?"
Kennett said, "Jim . . ."
One of the detectives handed Lucas a brown envelope. Lucas opened it and found a sheath of eight-by-ten color photos. Whitechurch, dead in the hallway, flat on his back. Blood on the tile behind his head, and on the wall. A twenty-dollar bill half pinned under the body.
"What's the money?" Lucas said.
"They must have been hassling over the cash when Bekker shot him," said the cop named Jim. "One of the janitors heard the shots. Not being stupid, he hollered before he went to look. Then he kind of carefully stuck his head through a fire door and saw Whitechurch on the ground. The outside door was just closing. Bekker must've grabbed what he could and run for it."
"He didn't take the eyelids," Lucas said. Except for the blood, Whitechurch might have been a sleeping drunk.
"Nope. Just poked him in the eyes and grabbed the dope, if there was any. They got a print, by the way, off a bill. It was Bekker."
"All right, let's get out there," Kennett said to the cops. There was an unhappy silence, all of them on their feet and moving through the door, shaking heads. "Hey. Everybody. Tell your people to put on the vests, huh? They're gonna be talking to some pissed-off people."
Huerta, bumping past Kennett, stopped to pat him on the head, pushing his hair down.
Kennett said, "What?" and Huerta, grinning, said, "Just knocking down your mohawk. With all that white hair stickin' up you looked like Steve Martin in _The Jerk,_ except skinny and old."
"Yeah, old, kiss my ass, Huerta," Kennett said, laughing, straightening his hair.
Lucas, astonished, watched Huerta walk away, then looked back at Kennett.
"What?" Kennett asked, puzzled, raking at his hair again.
"Steve Martin?" Lucas asked.
"Asshole," Kennett grumbled.
"They're probably calling you the same thing, you putting them on the street like that," Lucas said. Switching the topic away from Steve Martin, covering, covering . . .
"I know," Kennett said soberly, looking after the detectives. "Jesus, roustin' junkies in this heat . . . it's gonna stink and the junkies'll be pissed and the cops are gonna be pissed and somebody's gonna get hurt."
"Not a hell of a lot of choice," Lucas said. "Keep pushing everywhere. With Whitechurch dead, Bekker's gotta find a new source."
• • •
An hour later, Lucas lay on his bed at the Lakota and thought about what Huerta had said. That he looked like Steve Martin, with all that white hair . . .
All right. You're on the street. There's been a killing. A car speeds by and inside is an old white guy. That's what Cornell Reed told Bobby Rich's snitch. An old white guy. How would you know he was old, when he was in a moving car? If he had white hair . . .
And then there was Mrs. Logan, and what she'd said, in the apartment beneath Petty's . . . .
Kennett fit. He was a longtime intelligence operative. He was high up, with good access to inside information. He was tough but apparently well liked; he had charisma. He had white hair.
Kennett was sleeping with Lily. How did that cut across it? How did she wind up in the sack with a guy who might be a suspect? And the biggest question: with several hundred possible suspects, how did Kennett wind up in Lucas' lap, available for daily inspection?
O'Dell was one answer. Lily was another. Or both together.
He lay on the bed with the Magic Marker and his art pad, trying to put together a list. Finally he came up with:
1. Cornell Reed.
## CHAPTER
15
Lucas was flat on his back, half asleep, when Fell called. The room was semidark; he'd turned out all the lights but the one in the bathroom, and then half closed the door.
"I'm downstairs," she said. "If you're awake, let's get something to eat."
"Anything at Bellevue?" Lucas asked.
"I'll tell you about it."
"Ten minutes," he said.
He was fifteen minutes. He shaved, going easy over the bruises, brushed his teeth and took a quick shower, put on a fresh shirt, dabbed on after-shave. When he got down to the lobby, Fell looked him over and said, "Great. You make me feel like a rag."
"You look fine," he said, but she didn't. She looked worn, dirty around the eyes. The dress that had been crisp that morning hung slackly from her shoulders. "There's an Italian place a couple of blocks down that's friendly."
"Good. I couldn't handle anything complicated." As they were going out the door, she said, "I'm sorry about ditching you and going with Kennett, but this case really could mean a lot for me. And Mrs. Bedrick, she was mine . . . ours . . . and I wanted to be there to get the credit."
Lucas nodded and said, "No problem." On the sidewalk, he added, "You don't sound happy."
"I'm not. Bellevue's a rat's nest. They have a dial-in paging system, so now we're trying to figure out if we can match up the calls. And we're looking for people who might have been paging doctors who shouldn't have, that somebody else might have noticed. There are about two thousand suspects."
"Can you thin them out?"
"Maybe. We're trying extortion. Kennett worked out a routine with an assistant D.A. Everybody we talk to, we tell them the same thing: if we find out who Whitechurch's phone contact is before she comes forward, we'll charge her as an accomplice in the Bekker murders. If she comes forward and cooperates, we'll give her immunity on Bekker. And she can bring a lawyer and refuse to cooperate on anything else . . . . So there's a chance. If we can scare her enough."
"How do you know it's a _her?_ "
Fell grinned up at him: "That's Kennett. He said, 'Have you ever heard a male voice on a hospital intercom?' We all thought about it, and decided, Not very often. If a male voice kept calling out the names of nonexistent doctors—that's what we think she was doing, whoever she is, calling out code names—he'd be noticed. So we're pretty sure it's a her."
"What if it's just the switchboard?"
"Then we're fucked . . . although Carter thinks it probably isn't. A switchboard might start recognizing names and voices . . . ."
• • •
The Whetstone had an old-fashioned knife-grinding wheel in the window, a dozen tables in front, a few booths in back. Between the booths was a wooden floor, worn smooth and soft by a century of sliding feet. A couple turned slowly in the middle of it, dancing to a slow, sleepy jazz tune from an aging jukebox.
"Booth?" asked Lucas.
"Sure," said the waitress. "One left, in the no-smoking area."
Fell smiled ruefully at Lucas, and said, "We'll take it."
They ate spaghetti and garlic bread around a bottle of rosé, talking about Bekker. Lucas recounted the Minneapolis killings:
" . . . started killing them to establish their alibis. They apparently picked out the woman at the shopping mall at random. She was killed to confuse things."
"Like a bug. Stepped on," Fell said.
"Yeah. I once dealt with a sexual psychopath who killed a series of women, and I could understand him, in a way. He was nuts. He was _made_ nuts. If he'd had a choice, I'd bet that he'd have chosen not to be nuts. It was like, it wasn't his fault, his wires were bad. But with Bekker . . ."
"Still nuts," Fell said. "They might look cold and rational, but to be that cold, you've got to be goofy. And look what he's doing now. If we take him alive, there's a good chance that he'll be sent to a mental hospital, instead of a prison."
"I'd rather go to prison," Lucas said.
"Me, too, but there are people who don't think that way. Like doctors."
A heavyset man in work pants and a gray Charlie Chaplin mustache stepped across to the jukebox and stared into it. The waitress came by and said, "More wine?"
Lucas looked at Fell and then up at the waitress and said, "Mmmm," and the waitress took the glasses.
Behind her, the heavy man in work pants dropped a single quarter in the jukebox, carefully pressed two buttons, went back to his table and bent over the woman he had been sitting with. As she got up, the "Blue Skirt Waltz" began bubbling from the jukebox speakers.
"Jesus. _Blue Skirt._ And it's Frankie Yankovich, too," Lucas said. "C'mon, let's dance."
"You gotta be kidding . . . ."
"You don't want . . . ?"
"Of course I want," she said. "I just can't believe that you do."
They began turning around the floor, Fell light and delicate, a good dancer, Lucas denser, unskilled. They turned around the heavy man and his partner, the two couples caught by the same rhythm, weaving around the dance floor. The waitress, who'd taken menus to another table, lingered to watch them dance.
"One more time," the heavy man said to Lucas, in a heavy German accent, as the song ended. He bowed, gestured to the jukebox. Lucas dropped a quarter, punched "Blue Skirt," and they started again, turning around the tiny dance floor. Fell fit nicely just below his jaw, and her soft hair stroked his cheek. When the song ended, they both sighed and wandered back to the booth, holding hands.
"Sooner or later, I'd like to spend some time in your shorts, as we say around the Ninth," Fell said across the table as she sat down. "But not tonight. I'm too fuckin' dirty and miserable and tired and I've got too many bad movies in my head."
"Well," he said.
"Well, what? You don't want to?"
"I was thinking, well, I've got a shower."
She cocked her head, looking at him steadily, unsmiling. "You think it'll wash away that woman rolling over this morning, with those eyes?" she asked somberly.
After a moment, he said, "No. I guess not. But listen . . . you interest me. I think you knew that."
"I didn't really," she said, almost shyly. "I've got no self-confidence."
"Well." He laughed.
"You keep saying that. Well."
"Well. Have some more wine," he said.
Halfway through the second bottle of wine, Fell made Lucas play it again and they turned around the room, close, her face tipped up this time, breathing against his neck, warm, steamy. He began to react and was relieved to get her back to the booth.
She was drunk, laughing, and Lucas asked about the cop she used to date.
"Ah, God," she said, staring up at the ceiling, where a large wooden fan slowly turned its endless circles. "He was _so_ good-looking, and he was _such_ a snake. He used to be like this _Pope of Greenwich Village_ guy with these great suits and great shoes, and he hung out, you know? I mean, he was cool. His socks had clocks on them."
"How cool can a Traffic guy be?" Lucas cracked.
She frowned. "Were we talking about him? I don't . . ."
"Sure, at your place," he said, thinking, _As a matter of fact, you didn't, Lily did, Davenport, you asshole._ "I remember, mm, important details . . . ."
"Why's that important?" she asked, but she knew, and she was flattered.
"You're the fuckin' detective," Lucas said, grinning at her. "Have another drop of wine."
"Trying to get me drunk?"
"Maybe."
Fell put her wineglass on the table and poked a finger at him. "What the fuck are you doing, Davenport? _Are_ you Internal Affairs?"
"Jesus Christ—I told you, I'm not. Look, if you're really serious, my goddamn publisher's not far from here and my face is on the game boxes. There's a biography and everything, we could go over . . ."
"Okay. But why are you pumping me?"
"I'm not pumping you . . . ."
"Bullshit," she said. Her voice rose. "You're a goddamn trouser snake just like he was, and just like Kennett. I knew that as soon as you asked me to dance. I mean, I could feel myself _melting._ Now, what the fuck are you doing?"
Lucas leaned forward and said, trying to quiet her, trying not to laugh, "I'm not . . ."
"Jesus," she said, pulling back. She went back to the table and picked up her purse. "I'm really loaded."
"Where're we going?"
"Up to your room. I've changed my mind."
"Barbara . . ." Lucas threw three twenties at the tabletop, and hurried after her. "You're a little drunk . . ."
"Fuckin' trouser snake," Fell said as she led the way through the door.
He woke in the half-lit room, a thin arrow of light from the bathroom falling across the bed. He was confused, a feeling of déjà vu. Didn't Fell just call, didn't she say . . . ? He stopped, feeling the weight. She'd fallen asleep cradled beneath his arm, head on his chest, her leg across his right. He tried to ease out from beneath her, and she woke and said, "Hmmm?"
"Just trying to rearrange," he said, whispering, catching up with the night. She'd been almost timid. Not passive, but . . . wary.
"Um . . ." She propped herself up, her small breast peeking at him over the top of the blanket. "What time is it?"
Lucas found his travel clock, peered at it. "Ten minutes of three," he said.
"Oh, God." She pushed herself up, her back to him, and the sheet fell off. She had a wonderful back, he decided, smooth, slender, but with nice muscles. He drew a finger down her spine and she arched away from him. "Oooo. Stop that," she said over her shoulder.
"Come lay down," he said.
"Time to go."
"What?"
She turned to look at him, but her eyes were in shadow and he couldn't see them. "I really . . ."
"Bullshit. Come on and sleep with me."
"I really need some _sleep._ "
"So do I. Fuckin' Bekker."
"Forget Bekker for a few hours," she said.
"All right. But lay down."
She dropped back on the bed, beside him. "You're not still with Rothenburg?"
"No."
"It's over?"
"It's weird, is what it is," he said.
"You're not saying the right thing," said Fell. She propped herself up again, and he drew three fingers across the soft skin on the bottom of her breast.
"That's because Lily and I are seriously tangled up," Lucas said. "You know she's sleeping with Kennett."
"I figured. The first time I saw them together, she was dropping him off at Midtown South, and she kissed him good-bye and I had to go inside and put a cool wet rag on my forehead. I mean, hot. But then I saw you two talking to each other, you and Rothenburg, and it looked like unfinished business."
"Nah. But I was there when her marriage came apart and she helped kill off the last of my relationship with a woman I had a kid with. We were kind of . . . pivotal . . . for each other," Lucas said.
"All right," Fell said.
"Lily was driving?"
"What?"
"You said she dropped off Kennett."
"Well, yeah, Kennett can't drive. That'd kill him, the Manhattan traffic would." She sat up again, half turned, and this time he could see her eyes. "Davenport, what the _fuck_ are you up to?"
"Jesus . . ." He laughed, and caught her around the waist, and she let him pull her down.
"The one thing I want to know—if you're up to something, you're not screwing me to get it, are you?"
"Barbara . . ." Lucas rolled his eyes.
"All right. You'd lie to me anyway, so why do I ask?" Then she frowned and answered her own question: "I'll tell you why. Because I'm an idiot and I always ask. And the guys always lie to me. Jesus, I need a shrink. A shrink and a cigarette."
"So smoke, I don't mind," Lucas said. "Just don't dribble ashes on my chest."
"Really?" She scratched him on the breastbone.
"I mean, it's killing you, slowly but surely, but if you need one . . ."
"Thanks." She got out of bed—a wonderful back—found her purse, got her cigarettes, an ashtray and the TV remote. "I gotta get some nicotine into my bloodstream," she said. Ingenuously, genuinely, she added, "I didn't have a cigarette because I was afraid my mouth would taste like an ashtray."
"I thought you'd decided not to sleep with me, and changed your mind."
She shook her head. "Dummy," she said. She lit the cigarette and pointed the remote control at the TV, popped it on, thumbed through the channels until she got to the weather. "Hot and more hot," she said, after a minute.
"It's like Los Angeles, 'cept more humid," Lucas said.
"Shoulda been here last year . . . ."
They talked and she smoked, finished the cigarette, and then lit up another and went around the room and stole all his hotel matches. "I never have enough matches. I always steal them," she said. "When I'm working I've got two rules: pee whenever you can, and steal matches. No. Three rules . . ."
"Never eat at a place called Mom's?"
"No, but that's a good one," she said. "Nope: it's never sleep with a goddamn cop. Cops are so goddamn treacherous . . . ."
## CHAPTER
16
Sunday morning.
Sunlight poured like milk through the venetian blinds. Fell woke at nine o'clock, stirred, then half-sat, looking down at Lucas' dark head on the pillow. After a moment, she got up and stumbled around, picking up clothes. Lucas opened an eye and said, "Have I mentioned your ass?"
"Several times, and I appreciate all of them," she said. She offered a smile, but weakly. "My head . . . that goddamn cheap wine."
"That wine wasn't cheap." Lucas sat up, still sleepy, dropped his feet to the floor, rubbed the back of his neck. "I'll call Kennett, see if we can figure something out."
She nodded, still groggy. "I gotta go home to change clothes, then back to Bellevue. There'll be people around we wouldn't see during the week."
Lucas said, "This is really important to you, isn't it?"
"It's the biggest case I've ever been on," she said. "God, I'd love to get him. I mean, me, personally."
"You won't get him at Bellevue," Lucas said. "Even if you find Whitechurch's helper, and she talks, I wouldn't be surprised if Bekker's using a pay phone. Then where are you?"
"So if we find the phone, we can stake it out. Or maybe he uses one on the block where he lives, we can look at the apartments."
"Mmm."
"Maybe we'll get him tomorrow night, at the speech."
"Maybe . . . C'mon. I'll make sure you get clean in the shower."
"That's something I've always needed," she said. "Help in the shower."
"Well, you said your head feels weird. What you need is a hot shower and a neck massage. Really. I say this in a spirit of fraternity and sorority."
"Good, I don't think I could handle another sexual impulse," Fell said. But the shower took them back to the bed, and that took them back to the shower, and Fell was leaning against the wall, Lucas standing between her legs, drying her back with a rough terry-cloth towel, when Anderson called from Minneapolis.
"Cornell Reed. United to Atlanta out of La Guardia, transfer to Southeast to Charleston. No return. Paid for by the City of New York."
"No shit . . . Charleston?"
"Charleston."
"I owe you some bucks, Harmon," Lucas said. "I'll get back to you."
"No problem . . ."
Lucas hung up, turning it over in his head.
"What's Charleston?" Fell asked from the bathroom doorway.
"It's both a dance and a city . . . . Sorry, that was a personal call. I was trying to get through to my kid's mother. She's gone to Charleston with the Probe Team."
"Oh." Fell tossed the towel back into the bathroom. "You're still pretty tight with her?"
"No. We're done. Completely. But Sarah's my kid. I call her."
Fell shrugged and grinned. "Just checking the oil level," she said. "Are you going to call Kennett?"
"Yeah."
They ate a quick breakfast in the hotel coffee shop, then Lucas put Fell in a cab back to her apartment. He called Kennett from his room and got switched from Midtown South to a second phone. Kennett picked it up on the first ring.
"If we don't get him tomorrow, at the speech, I'm heading back to the Twin Cities, see what I can find," Lucas said.
"Good. I think we've got all the routine stuff pinned down here," Kennett said. "Lily's here, and we were about to call you. We're thinking about a boat ride."
"Where's here?" Lucas asked.
"Her place."
"So come and get me," Lucas said.
After talking to Kennett, Lucas sat with his hand on the phone, thinking about it, then picked it up again, dialed the operator, and got the area code for Charleston. He had no idea how big the city was, but had the impression that it was fairly small. If they knew assholes in Charleston the way they knew them in the Twin Cities . . .
The information service got him the phone number for the Charleston police headquarters, and two minutes later, he had the weekend duty officer on the line.
"My name is Lucas Davenport. I'm a cop working out of Midtown South in Manhattan. I'm looking for a guy down your way, and I was wondering about the prospects of finding him."
"What's the problem?" A dry southern drawl, closer to Texan than the mush-mouth of South Carolina.
"He saw a guy get shot. He didn't do it, just saw it. I need to talk to him."
"What's his name?"
"Cornell Reed, nickname Red. About twenty-two, twenty-three . . ."
"Black guy." It was barely a question.
"Yeah."
"And you're from Midtown South."
"Yeah."
"Hang on . . ."
Lucas was put on hold, waited for a minute, then two. Always like this with cops. Always. Then a couple of clicks, and the line was live again. "I got Darius Pike on the line, he's one of our detectives . . . . Darius, go ahead . . ."
"Yeah?" Pike's voice was deep, cool. Children were laughing in the background. Lucas identified himself again.
"Am I getting you at home? I'm sorry about that . . . ."
" 'S okay. You're looking for Red Reed?"
"Yeah. He supposedly witnessed a killing up here, and I'm pretty hot to talk to him."
"He came back to town a month ago, the sorry-ass fool. You need to bust him?"
"No, just talk."
"Want to come down, or on the phone?"
"Face-to-face, if I can."
"Give me a call a day ahead. I can put my hands on him about any time."
• • •
Now he had to make a decision: Minneapolis, Charleston. Two different cases, two different leads. Which first? He thought about it. He wouldn't be able to get down to Charleston and back in time. The New School trap was the next night; if they didn't get Bekker, then the trip to Minneapolis was critical. Bekker was killing people, after all. Charleston might shed some light on Robin Hood, and Robin Hood was killing people, too—but those were mostly _bad_ people, weren't they? He shook his head wryly. It wasn't supposed to matter, was it? But it did.
Lucas made one more call, to Northwest Airlines, and got a seat to Minneapolis-St. Paul, then a triple play, Minneapolis-St. Paul to Charleston to New York. There, that was all he could do for now. It all hinged on tomorrow night.
When Lily called from the front desk, he'd changed to jeans and blue T-shirt. He went down, found her waiting, eyes tired but relaxed. She was wearing jeans and a horizontally-striped French fisherman's shirt that might have cost two hundred dollars on Fifth Avenue, and an aqua-colored billed hat.
"You look like a model," he said.
"Maybe I oughta call _Cruising World._ "
"Yeah, you look kinda gay," he said.
"That's a sailboat magazine, you dope," she said, taking a mock swipe at him.
Kennett was waiting in the passenger seat of a double-parked Mazda Navaho, wearing comfortable old khakis and a SoHo Surplus T-shirt.
"Nice truck," Lucas said to Lily as he crawled in back.
"Kennett's. Four-wheel drive must help testosterone production," Lily said, walking around to the driver's side and climbing in. "You've got one, don't you?"
"Not like this: this is sort of a _Manhattan_ four-wheel drive," he said, tongue in cheek. To Kennett he said, "I didn't think you could drive."
"Got it before the last attack," Kennett said. "I think the price is what brought the attack on. And don't give me any shit about Manhattan four-by-fours, this is a fuckin' workhorse . . . ."
"Yeah, yeah . . ."
They left Manhattan through the Lincoln Tunnel, emerging in Jersey, took a right and then followed a bewildering zigzag path back to the waterfront. The marina was a modest affair, filling a dent in the riverbank, a few dozen boats separated from a parking lot by a ten-foot chain-link fence topped with razor wire. Most of the boats were in concrete slips, halyards clinking softly against the aluminum masts like a forest of one-note wind chimes; a few more boats were anchored just offshore.
"Look at this guy, putting up his 'chute," Kennett said, climbing down from the truck. Lucas squeezed out behind him as Lily climbed out of the driver's seat. Kennett pointed out toward the river, where two sailboats were tacking side-by-side down the Hudson, running in front of a steady northwest breeze, their sails tight with the wind. A man was standing on the foredeck of one of them, freeing a garish crimson-and-yellow sail. It filled like a parachute, and the boat leapt ahead.
"You ever sailed?" Kennett asked.
"A couple times, on Superior," Lucas said, shading his eyes. "You feel like you're on a runaway locomotive. It's hard to believe they're barely going as fast as a man can jog."
"A man doesn't weigh twenty thousand pounds like that thing," Kennett said, watching the lead boat. "That _is_ a locomotive . . . ."
They unloaded a cooler from the back of the truck and Lucas carried it across the parking lot, past a suntanned woman in a string bikini with a string of little girls behind her, like ducklings. The smallest of the kids, a tiny red-headed girl with a sandy butt and bare feet, squealed and danced on the hot tarmac while carrying a pair of flip-flops in her hands.
Lily led the way through a narrow gate in the chain-link fence, Lucas right behind her, Kennett taking it slow, down to the water. Here and there, people were working on their boats, listening to radios as they worked. Most of the radios were tuned to rock stations, but not the same ones, and an aural rock-'n'-roll fest played pleasantly through the marina. Few of the boats actually seemed ready to go out, and the work was slow and social.
"There she blows, so to speak," Kennett said. The _Lestrade_ was fat and graceful at the same time, like an overweight ballerina.
"Nice," Lucas said, uncertainly. He knew open fishing boats, but almost nothing about sailboats.
"Island Packet 28—it _is_ a nice boat," Kennett said. "I got it instead of kids."
"Not too late for kids," Lucas said. "I just had one myself."
"Wait, wait, wait." Lily laughed. "I should have a say in this."
"Not necessarily," Lucas said. He stepped carefully into the cockpit, balancing the cooler. "The goddamned town is overrun with nubile prospects. Find somebody with a nice set of knockers, you know, not too smart so you wouldn't have to worry about the competition. Maybe with a fetish for housework . . ."
"Fuck the sailing, let's go back into town," Kennett said.
"God, I'm looking forward to this," Lily said. "The flashing wit, the literary talk . . ."
Lily and Lucas rigged the sails, with Kennett impatiently supervising. When he was bringing the sails up, Lucas took a moment to look through the boat: a big berth at the bow, a tidy, efficient galley, a lot of obviously custom-built bookshelves jammed with books. Even a portable phone.
"You could live here," Lucas said to Kennett.
"I do, a lot of the time," Kennett said. "I probably spend a hundred nights a year on the boat. Even when I can't sail it, I just come over here and sit and read and sleep. Sleep like a baby."
Kennett took the boat out on the motor, his fine white hair standing up like a sail, his eyes shaded by dark oval sunglasses. A smile grew on his tanned face as he maneuvered out along the jetty, then swung into the open river. "Jesus, I love it," he said.
"You gotta be careful," Lily said anxiously, watching him.
"Yeah, yeah, this takes two fingers . . . ." To Lucas he said, "Don't have a heart attack—it just unbelievably fucks you up. I can run the engine and steer, but I can't do anything with the sails, or the anchor. I can't go out alone."
"I don't want to talk about it," Lucas said.
"Yeah, fuck it," Kennett agreed.
"What does it feel like?" Lucas asked.
"You weren't gonna talk about it," Lily protested.
"It feels like a pro wrestler is trying to crush your chest. It hurts, but I don't remember that so much. I just remember feeling like I was stuck in a car-crusher and my chest was caving in. And I was sweating, I remember being down on the ground, on the floor, sweating like a sonofabitch . . . ." He said it quietly, calmly enough, but with a measure of hate in his voice, like a man swearing revenge. After another second, he said, "Let's get the sails up."
"Yeah," Lucas said, slightly shaken. "I gotta pull on a rope, right?"
Kennett looked at the sky. "God, if you heard the man, forgive him, the poor fucker's from Minnesota or Missouri or Montana, some dry-ass place like that."
Lucas got the mainsail up. The jib was on a roller, with the lines led back to the cockpit. Lily worked it from there, sometimes on her own, sometimes with prodding from Kennett.
"How long have you been sailing?" Lucas asked her.
"I did it when I was a kid, at summer camp. And then Dick's been teaching me the big boat."
"She learns quick," Kennett said. "She's got a natural sense for the wind."
They slid lazily back and forth across the river, water rushing beneath the bow, wind in their faces. A hatch of flies was coming off the water, their lacy wings delicately floating around them. "Now what?" Lucas asked.
Kennett laughed. "Now we sail up and then we turn around, and sail back."
"That's what I thought," Lucas said. "You're not even trolling anything."
"You're obviously not into the great roundness of the universe," Kennett said. "You need a beer."
Kennett and Lily gave him a sailing lesson, taught him the names of the lines and the wire rigging, pointed out the buoys marking the channel.
"You've got a cabin on a lake, right? Don't you have buoys?"
"On my lake? If I peed off the end of the dock, I'd hit the other side. If we put in a buoy, we wouldn't have room for a boat."
"I thought the great North Woods . . ." Kennett prompted, seriously.
"There's some big water," Lucas admitted. "Superior: Superior'll show you things the Atlantic can't . . . ."
"I _seriously_ doubt that," Lily said skeptically.
"Yeah? Well, once every few years it freezes over—and you look out there, a horizon like a knife and it's ice all the way out. You can walk out to the horizon and never get there . . . ."
"All right," she said.
They talked about ice-boating and para-skiing, and always came back to sailing. "I was planning to take a year off and single-hand around the world, maybe . . . unless I got stuck in the Islands," Kennett said. "Maybe I would have got stuck, maybe not. I took Spanish lessons, took some French . . ."
"French?"
"Yeah . . . you run down the Atlantic, see, to the Islands, then across to the Canaries, maybe zip into the Med for a look at the Riviera—that's French—then come back out and down along the African coast to Cape Town, then Australia, then Polynesia. Tahiti: they speak French. Then back up to the Galápagos, Colombia and Panama, and the Islands again . . ."
"Islands—I like the idea," Lucas said.
"You like it?" asked Kennett, seriously.
"Yeah, I do," Lucas said, looking out across the water. His cheekbones and lips were tingling from the sun, and he could feel the muscles relax in his neck and back. "I had a bad time a year ago, a depression. The medical kind. I'm out now, but I never want to do that again. I'd rather . . . run. Like to the Islands. I don't think you'd get depressed in the Islands."
"Exactly what islands are we talking about?" Lily asked.
"I don't know," Kennett said vaguely. "The Windwards, or the Leewards, or some shit . . ."
"What difference would it make?" Lucas asked Lily.
She shrugged: "Don't ask me, they're your islands."
After a moment of silence, Kennett said, "A unipolar depression. Did you hear your guns calling you?"
Lucas, startled, looked at him. "You've had one?"
"Right after the second heart attack," Kennett said. "The second heart attack wasn't so bad. The depression goddamned near killed me."
They turned and started back downriver. Kennett fished in his pocket and pulled out a pack of cigarettes.
"Dick. Throw those fuckin' cigarettes . . ."
"Lily . . . I'm smoking one. Just one. That's all for today."
"God damn it, Dick . . ." Lily looked as though she were going to cry.
"Lily . . . aw, fuck it," Kennett said, and he flipped the pack of Marlboros over the side, where they floated away on the river.
"That's better," Lily said, but tears ran down her cheeks.
"I tried to bum one from Fell the other day, but she wouldn't give it to me," Kennett said.
"Good for her," said Lily, still teary-eyed.
"Look at the city," Lucas said, embarrassed. Kennett and Lily both turned to look at the sunlight breaking over the towers in Midtown. The stone buildings glowed like butter, the modern glass towers flickering like knives.
"What a place," Kennett said. Lily wiped her cheeks with the heels of her hands and tried to smile.
"Can't see the patches from here," Lucas said. "That's what New York is, you know. About a billion patches. Patches on patches. I was walking to Midtown South from the hotel, crossing Broadway there at Thirty-fifth, and there was a pothole, and in the bottom of the pothole was another pothole, but somebody had patched the bottom pothole. Not the big one, just the little one in the bottom."
"Fuckin' rube," Kennett muttered.
They brought the boat back late in the afternoon, their faces flushed with the sun. And after Lucas dropped the mainsail, Lily ran it into the marina with a soft, skillful touch.
"This has been the best day of my month," Kennett said. He looked at Lucas. "I'd like to do it again before you go."
"So would I," Lucas said. "We oughta go down to the Islands sometime . . . ."
Lucas hauled the cooler back to the truck and Lily brought along an armload of bedding that Kennett wanted to wash at home.
"Shame that he can't drive the truck," Lucas said as Lily popped up the back lid.
"He does," she said in a confidential voice. "He tells me he doesn't, but I know goddamn well that he sneaks out at night and drives. A couple of months ago I drove back to his place, and when we parked I noticed that the mileage was something like 1-2-3-4-4, and I was thinking that if I only drove one more mile, I'd have a straight line of numbers: 1-2-3-4-5. When I came over the next day, the mileage was like 1-2-4-1-0, or something like that. So he'd been out driving. I check it now, and lots of times the mileage is up. He doesn't know . . . . I haven't mentioned it, because he gets so pissed. I'm afraid he'll get so pissed he'll have another attack. As long as it has power steering and brakes . . ."
"It'll drive a guy nuts, being penned up," Lucas said. "You oughta stay off his case."
"I try," she said. "But sometimes I just can't help it. Men can be so fucking stupid, it gives me a headache."
They went back to the boat and found Kennett below, digging around. "Hey, Lucas, a little help? I need to pull this marine battery, but it's too heavy for Lily."
"Dick, are you messing around with that wrench again . . . ?" Lily started, but Lucas put an index finger over his lips and she stopped.
"I'll be down," Lucas said.
Ten minutes later, while Kennett and Lily did the last of the buttoning-up, Lucas humped the battery back to the car. In the parking lot, he propped one end of it on the truck bumper while he sorted out the keys, then turned and looked back through the fence. Lily and Kennett were on the dock, Lily leaning into him, his arms around her waist. She was talking to him, then leaned forward and kissed him on the mouth. Lucas felt a pang, but only a small one.
Kennett was okay.
## CHAPTER
17
The New School auditorium was compact, with a narrow lobby between the interior auditorium doors and the doors to the street.
"Perfect," Lucas told Fell. They'd taken the tour with a half-dozen other cops, and now, waiting, wandered outside to Twelfth Street. Fell lit a cigarette. "Once he comes around the corner, he'll be inside the net. And the lobby's small enough that we can check everyone coming through before they realize there are cops all over the place."
"You still think he'll show?" Fell asked skeptically.
"Hope so."
"It'd be too easy," she said.
"He's a nut case," Lucas said. "If he's seen the announcement, he'll be here."
A car dropped Kennett at the curb. "Opening night," he said as he climbed out. He looked up and down the fashionable residential street, bikes chained to wrought-iron fences, well-kept brick townhouses climbing up from the street. "It feels like something's gonna happen."
They followed him inside, and Carter came by with radios. They each took one, fitting the earpieces, checking them out. "Stay off unless it's critical," Carter said. "There are twelve guys here, and if all twelve start yelling at the same time . . ."
"Where do you want me?" Lucas asked.
"Where do you think?" Carter asked. "Ticket booth?"
"Mmm, I'd be looking at too many people's backs," Lucas said. He glanced around. A short hall led from the auditorium lobby to the main entrance lobby of the New School. "How about if I stood back there in the hall?"
"All right," Carter said. To Fell, he said, "We've got you handing out programs. You'll be right there in the lobby."
"Terrific . . ."
"What's the setup?" Kennett asked.
"Well, we're supposed to start in twenty minutes. We've got you just inside the auditorium entrance, where you can see everyone, or get back out to the lobby in a hurry," Carter said. "It's right down here . . . ."
Bekker tottered down Twelfth Street ten minutes before the lecture was scheduled to begin, past a guy working on a car in the failing daylight. Bekker was nervous as a cat, excited, checking the scattering of people walking along the street with him, and toward him, converging on the auditorium. This was dangerous. He could feel it. They'd be talking about him. There might be cops in the crowd. But still: worth it. Worth some risk.
Most of the people were going through a series of theater-style doors farther up the street. That would be the auditorium. There was another door, closer. On impulse he entered there, turned toward the auditorium.
Almost stumbled.
_Davenport._
Trap.
The fear almost choked him, and he caught at his throat. Davenport and another man, their backs to Bekker, were in the hallway between the separate entries. Not ten feet away. Watching the crowd come through the other door.
Davenport was to the left, half turned toward the second man, his back directly to Bekker. The second man, half turned toward Davenport, glanced toward Bekker as simple momentum took Bekker inside. Couldn't stop. He went straight through the school lobby, past the entrance to the auditorium. An empty guard desk was to the right, with a phone behind it. Ahead of him, another hallway that seemed to lead back outside.
Bekker unconsciously touched his face, felt the hard scars under the special makeup. That night in the funeral home, Davenport hacking at him . . .
Bekker wrenched himself back, forced himself to walk down the stairs, through the next door, outside. He was sweating, almost gasping for breath.
He found himself in a sculpture garden, facing another door like the one he'd come through. On the other side of the door was a hallway, and beyond that, maybe a hundred feet away, another set of doors and the next street. Nobody ahead. He strode quickly across the courtyard, caught the door, pulled.
Locked. Stricken, he gave it a tug. It didn't budge. The glass was too thick to break, even if he had something to break it with. He turned and looked back, toward the way he'd come. If he tried to get out that way, he'd be face to face with Davenport for several seconds, just as he'd been with the cop Davenport had been talking to.
He stood, frozen, unable to sort the possibilities. He had to get out of sight. He went to his left, found a short hallway with a door marked with a B and the word "Stair." He jerked at the door, hoping . . .
Locked. Damn. He huddled in the doorway, temporarily out of sight. But he couldn't stay: if anybody saw him like this, hiding, they'd know.
Another goddamned Davenport trap, pulling him in . . .
Bekker lost it for a moment, his mind going away, dwindling, imploding . . . . He came back with a gasp, found himself pulling at the door, fighting the door handle.
No. There must be something else. He let go of the door, turned back to the courtyard. He needed help, needed to think. He groped for his pillbox, found it, gulped a half-dozen crosses. The acrid taste on his tongue helped cool him, get him thinking again.
If they caught him—and if they didn't kill him—they'd put him back inside, they'd pull him off his chemicals. Bekker shuddered, a full-body spasm. Take him off: he couldn't live through that again, he couldn't even think about it.
He thought of the funeral home again. Davenport's face, inches from his, screaming, the words unintelligible, then the pistol coming up, the gunsight coming around like a nail on a club, the nail ripping through his face . . .
Had to think. Had to think.
Had to move. But where? Davenport was right there, watching. Had to get past him. Only half aware of what he was doing, he fetched the pill box and gulped the rest of the speed and a single tab of PCP. Think.
"They gotta start pretty soon," Carter said.
"Give him another five minutes," Davenport said. "Fuck around with the slide projector or something."
"The crowd's gonna be pissed when Yonel makes the announcement."
"Maybe not," said Kennett, who'd gotten tired of waiting in the auditorium. "Maybe they'll get a kick out of it."
"Yonel says he'll do a half-hour on Mengele and Bekker anyway, before he says anything," Lucas said. He stood and stepped to the door: "I'm going to take a quick turn through the crowd. There're not many people coming in."
"Fuck it, he's not coming," Carter said.
"Maybe not, but he should have," Lucas said.
Bekker, desperately exploring the courtyard, followed a short flight of steps into an alcove and found another door. Behind the stage? Would there be cops back there? He took the handle in his hand, pulled . . . and the door moved. He eased it open until just a crack of light was visible and pressed his eye to the opening. Yes. Backstage. A man was there, wearing slacks and a sport coat, peering out at the audience from a dark corner on the opposite side of the stage. As Bekker watched, he lifted a rectangular object to his face. A radio? Must be. Cop.
Just inside the door, in front of Bekker, was a scarred table, and on the table an empty peanut butter jar, a black telephone and what looked like a collapsible umbrella in a nylon case. Bekker let the door close, turned back toward the steps. A finger of despair touched him: no way out. No way. And they'd be checking the building before they left. He knew that. He had to get out. Or hide.
Wait. A radio? The cop had a _radio._
Bekker turned, went back to the door, peeked inside again. The cop was still in the corner, peering out from behind the curtain, checking the crowd. And on the table, not an umbrella, but a folding music stand, apparently left behind after a concert.
He flashed on Ray Shaltie, and the blood splashing from his head . . . .
The PCP was coming up now, warming him, bringing him confidence. He needed that radio. He let the door close, took a quick, silent turn around the alcove outside the door, thinking. A paper? He dug in his bag, found an envelope, folded it. Thought again for a moment, but there was no other way: he _would not_ be beaten. Bekker took a breath, posed for a moment, then stepped to the door, pulled it open, and stepped inside.
The cop saw him immediately and frowned, took a step toward him. Bekker held up the envelope, and in a whisper, called, "Officer. Officer."
The cop glanced out at the crowd, then started across the stage behind the curtain. Radio in his hand. Bekker took a step forward, touched the music stand. It would be flimsy when opened, but when closed, and wrapped in its plastic sheath, a perfect club.
"You're not . . ." the cop started. Deep voice.
"The man out there . . ." Bekker whispered, and thrust the envelope at the cop, dropping it at the same time. The envelope fell to the cop's feet. Without thinking, the cop bent to catch it.
And Bekker hit him.
Hit him behind the ear with the music stand, swinging it like a hatchet. The impact sounded like a hammer striking an overripe cantaloupe, and the cop went down, the radio hitting the floor beside him. There'd been little noise, and that was muffled by the curtains, Bekker thought, but he hooked the man by the collar and dragged him into the corner by the door. And waited. Waited for the call, for the shout, that would end it. Nothing.
The cop couldn't be allowed to talk about how he was ambushed. Bekker stood over him for a moment, waiting, waiting, then pushed open the exterior door, dragged the body through it. The courtyard was still empty. Bekker lifted the music stand and hit the unconscious cop again and again, until the head resembled a bloody bag of rice.
_Stop . . . no time. But the eyes . . ._
Hurrying now, he used his penknife to cut the eyes, then patted down the body and found an identification card: Francis Sowith. The radio. Shit. The radio was still inside. He went to the door, peeked through, saw the radio, stepped quickly inside and retrieved it.
Back out on the porch again, stepping over the dead man. He noticed he had blood on his hands, and wiped them on the cop's coat. Still sticky: he lifted them to his face and sniffed. The smell of the blood was familiar, comforting.
He looked at the radio. Basic thumb switch. Calmed himself, checked his clothing, straightened it, and walked up the steps to the door back inside.
He took a breath, tensing, opened the door, and walked straight ahead. A staff member, he thought. That's what he was: a teacher who worked here. He heard a voice, a man, from around the corner. He slipped up to the guard desk, where he'd seen the telephone, and stepped around behind the desk, the phone to his ear. He could see the shoulder and sleeve of Davenport's jacket now, if that was in fact Davenport, in the same place. He leaned over the desk, head down, put the radio to his mouth, and thumbed the switch.
"This is Frank," he blurted. "He's here, backstage, backstage . . . ."
He dropped the radio hand, and pressed the phone receiver to his ear, his shoulder turned away: the body language said _making a date._ At the same time, there was a shout, then another. Davenport's shoulder disappeared from the doorway, but another man came through it, running, right past the desk and down into the courtyard.
Moving quickly, Bekker walked from behind the desk, looking straight ahead, out through the school doors into the street. A woman screamed from the auditorium. Bekker kept walking. The man who'd been working on the car hurried past him, heading toward the doors, a pistol in his hand.
And then the night closed around him. Bekker was gone.
## CHAPTER
18
They wound up in the courtyard, a half-dozen senior police officers shouting at each other. Lights burned in every room of the building and uniformed cops crawled through it inch by inch, but the people in the courtyard knew the search was pointless.
"Silly motherfucker . . . How many got out? How many?"
"I was trying to save his ass. Where the fuck were your guys, huh? Where the fuck . . ." A square guy pushed a tall guy, and for a moment it looked like a fight; but then other cops got between them.
"Jesus Christ, you gotta go out the back, the fuckin' TV is sweeping the streets . . . ."
"Who had the watch on the stairs? Where was . . ."
"Shut up." Kennett had been sitting on a bench, talking to Lily and O'Dell. Now he shouldered through the ring of cops, his voice cutting through the babble like an icicle going through a sponge cake. "Shut the fuck up."
He stood on the sidewalk, pale, two fingers hovering over his heart. He turned to one of the cops: "How many got out?"
"Listen, it wasn't my . . ."
"I don't give a shit whose fault it was," Kennett snarled. "We all fuckin' blew it. What I want to know is, how many got out?"
"I don't know," the cop said. "Twenty or thirty. When everybody stampeded backstage, a bunch of people in the lobby and near the doors just went outside. Nobody was there to stop them. When I came back . . . most of them were gone."
"There were only about fifty people in the auditorium," Kennett said. "So maybe half of them got out."
"But that's not the thing," the cop said.
"What's the thing?" Kennett asked. His voice was like a hangnail, sharp, ragged, painful.
"The thing is, I looked into every one of those faces. Bekker wasn't there. I don't care if you hang me up by my nuts, you ain't gonna get me to say he was, 'cause he wasn't. He wasn't there."
"He had to be somewhere," Carter snapped.
"Nobody came across the stage. Nobody went out through the courtyard. There was only one other door, and that doesn't go anywhere, it just comes back to the lobby . . . ."
There was a long moment of silence, compounded of anger and fear. Heads would roll for this one. Heads would roll. A couple of cops glanced furtively at O'Dell and Lily, deep in private talk. After a moment, Huerta said, "He must've been here all the time. He must've hid out before we got here, saw that he couldn't get out, figured we'd sweep the place before we left, and nailed Frank to get his radio."
Kennett was nodding. "That couldn't have been Frank who called . . . ."
"Sounded like Frank . . . ."
"So Bekker's got a deep voice, big fuckin' deal. We had people back there in five seconds, and Frank was gone. It took a while to mess him up like that."
"Then why'd he call? Bekker? If he was already gone?" Kuhn asked.
"To get us running back there," Lucas said. "Say he goes back there, nails Frank, takes the radio, goes off through the side door around the corner from the lobby, makes the call, then pushes through the door and goes right through the lobby and out."
"Billy said nobody came through the door," Kuhn said.
A young plainclothes cop with his hands in his pockets shook his head. "I swear to God, I don't see how anybody could've got through there. Lieutenant Carter told me to stay there, and even when Frank called, I stayed there. I saw everybody running . . ."
"But your back was to the door?" Kennett asked.
"Yeah, but I was right _there,"_ the young cop said. He could feel the goat horns being fitted for his head.
Kennett turned to Lucas: "You're sure he didn't come past you?"
"I don't see how. It's like this guy said . . ." Lucas pointed at the cop who looked at the faces. "I looked at every goddamn face coming through the door; he just wasn't there."
"All right, so he was inside," Kennett said. "We assume he made the radio call as a diversion to get out . . . ."
"Or to hide," somebody said. "If he had a bolthole during the day . . ."
"We'll find out," Kennett said, peering up at the brightly lighted windows. He glanced sideways at Lucas, who shook his head. Bekker was gone. "The other possibility is that he went out a window somewhere and made the radio call to pull the guys off the street . . . ."
"What if he had keys and was already outside, and was just taunting us?" one of the cops asked.
They talked for twenty minutes before drifting away to specific assignments, or simply drifting away, afraid that their names and faces might become associated with the disaster. In the alcove outside the stage door, a crime-scene crew worked under heavy lights, picking up what they could. But there was no real question: it was Bekker. But Bekker, how?
"Okay, now we're out of cop work: now we're down to politics," Kennett said to Lucas as they stood together in the courtyard.
"You gonna hang?" Lucas asked.
"I could," Kennett nodded. "I gotta start calling people, gotta get some spin on the thing, fuzz it up."
"Gonna be tough, with you right here," Lucas said.
"So what would you do?" Kennett asked.
"Lie," Lucas said.
Kennett was interested. "How?"
"Blame Frank. Unlock the back door," Lucas said, nodding to the opposite side of the courtyard. "Tell them that Bekker hid in the building during the day and that he must've stolen keys from somewhere. That when he came out and got down here, cutting through the courtyard, using his keys—where we only had one man, because we'd secured the place ahead of time—he ran head-on into Frank. There was a fight, but Bekker's a PCP freak and he killed Frank and escaped back out the other side of the building. If anybody gets blamed, the blame goes on Frank. But nobody'll say anything, because Frank's dead. You could even do a little off-the-record action. _Tell_ them that Frank fucked up, but we can't say it publicly. He was a good guy and now he's dead . . . ."
"Hmph." Kennett pulled at his lip. "What about the radio call?"
"Somebody's already suggested that he was taunting us: go with that," Lucas suggested. "That he was already outside. That fits Bekker's character, as far as the media's concerned."
"Do you think . . . ?"
"No, I think he suckered us."
"So do I." Kennett stared at his feet for a moment, then glanced at Lily and O'Dell. "The story might not hold up for long."
"If we get him before it breaks, nobody'll care."
Kennett nodded. "I better go talk to O'Dell. We'll need a ferocious off-the-record media massage."
"You think he'll help?"
Kennett permitted himself a very thin grin. "He was here too," Kennett said. "They'd just pulled up outside . . ."
Kennett started toward Lily and O'Dell, then stopped and turned, hands in his pockets, no longer grinning. "Get your ass back to Minneapolis. Find something for us, God damn it."
## CHAPTER
19
Lucas sat alone in the worst row of seats on the plane, in tourist class behind the bulkhead, no good place to put his feet except in the aisle. The stewardess was watching him before they crossed Niagara Falls.
"Are you all right?" she asked finally, touching his shoulder. He'd dropped the seat all the way back, tense, his eyes closed, like a patient waiting for a root canal.
"Are the wheels off the ground?" he grated.
"Uh-oh," she said, fighting a smile. "How about a scotch? Double scotch?"
"Doesn't work," Lucas said. "Unless you've got about nine phenobarbitals to put in it."
"Sorry," she said. Her face was professionally straight, but she was amused. "It's only two more hours . . . ."
"Wonderful . . ."
He could see it so clearly in his mind's eye: ripped chunks of aluminum skin and pieces of engine nacelle scattered around a Canadian cornfield, heads and arms and fingers like bits of trash, fires guttering just out of sight, putting out gouts of oily black smoke; women in stretch pants wandering through the wreckage, picking up money. A Raggedy Ann doll, cut in half, smiling senselessly; all images from movies, he thought. He'd never actually seen a plane crash, but you had to be a complete idiot not to be able to imagine it.
He sat and sweated, sat and sweated, until the stewardess came back and said, "Almost there."
"How long?" he croaked.
"Less than an hour . . ."
"Sweet bleedin' Jesus . . ." He'd been praying that it was only a minute or two; he'd been sure of it.
The plane came in over the grid of orange sodium-vapor lights and blue mercury lights, banking, Lucas holding on to the seat. The window was filled with the streaming cars, the black holes of the lakes stretching down from just west of the Minneapolis Loop. He looked at the floor. Jumped when the wheels came down. Made the mistake of glancing across the empty seat next to him and out the window, and saw the ground coming and closed his eyes again, braced for the impact.
The landing was routine. The bored pilot said the usual good-byes, the voice of a Tennessee hay-shaker, which he undoubtedly was, not qualified to fly a '52 Chevy much less a jetliner . . . .
Lucas stunk with fear, he thought as he bolted from the plane, carrying his overnight bag. _My God, that ride was the worst._ He'd read that La Guardia was overcrowded, that in a plane you could get cut in half in an instant, right on the ground. And he'd have to do it again in a day or two.
He caught a cab, gave directions, collapsed in the backseat. The driver took his time, loafing along the river, north past the Ford plant. Lucas' house had a light in the window. The timer.
"Nice to get home, huh?" the cabdriver asked, making a notation in a trip log.
"You don't know how good," Lucas said. He thrust a ten at the driver and hopped out. A couple strolled by on the river walk, across the street.
"Hey, Lucas," the man called.
"Hey, Rick, Stephanie." Neighbors: he could see her blond hair, his chrome-rimmed glasses
"You left your backyard sprinkler on. We turned it off and put the hose behind the garage."
"Thanks . . ."
He picked up the mail inside the door, sorted out the ads and catalogs and dumped them in a wastebasket, showered to get the fear-stink off his body and fell into bed. In thirty seconds, he was gone.
"Lucas?" Quentin Daniel stuck his head out of his office. He had dark circles under his eyes and he'd lost weight. He'd been the Minneapolis chief of police for two terms, but that wasn't what was eating him. Innocent people had died because of Quentin Daniel: Daniel was a criminal, but nobody knew except Daniel and Lucas. Lucas had resolved it in his mind, had forgiven him. Daniel never could . . . . "C'mon in. What happened to your face?"
"Got mugged, more or less . . . I need some help," Lucas said briefly, settling into the visitor's chair. "You know I'm working in New York."
"Yeah, they called me. I told them you were Mr. Wunnerful."
"I need to find the guys who were in the jail cells next to Bekker—or anybody he talked to while he was in there."
"Sounds like you're scraping the bottom of the bucket," Daniel said, playing with a humidor on his desk.
"That's why I'm here," Lucas said. "The cocksucker's dug in, and we can't get him out."
"All right." Daniel picked up his phone, punched a number. "Is Sloan there? Get him down to my office, will you? Thanks."
There was a moment of awkward silence, then Lucas said, "You look like shit."
"I feel like shit," Daniel said. He turned the humidor around, squared it with the edge of the desk.
"Your wife . . . ?"
"Gone. Thought it'd be a lift, seeing her go, but it wasn't. I'd get up every morning and look down at her and wish she was gone, and now I get up and look at the bed and there's a hole in it."
"Want her back?"
"No. But I want something, and I can't have it. I'll tell you one thing, between you and me and the wall—I'm getting out of here. Two months and I hit a crick in the retirement scale. Maybe go up north, get a place on a lake. I've got the bucks."
There was a knock on the door, and Daniel's secretary stuck her head in and said, "Sloan . . ."
Lucas stood up. "I do wish you luck," he said. "I'm serious."
"Thanks, but I'm cursed," Daniel said.
Sloan was lounging in the outer office, a cotton sport coat over a tennis shirt, chinos, walking shoes. He saw Lucas and a grin spread across his thin face.
"Are you back?" he asked, sticking out his hand.
Lucas, laughing: "Just for the day. I gotta find some assholes and I need somebody with a badge."
"You're working in the Big Apple . . . ."
"Yeah. I'll tell you about it, but we gotta go talk to the sheriff."
Three names, a deputy sheriff said. He'd looked at the records, checked with the other guards. They all agreed.
Bekker had been next to Clyde Payton, who was now at Stillwater, doing twenty-four months on a drugstore burglary, third offense. A doper.
"Motherfucker's gonna come out and kill people," the deputy said. "He thought Bekker was like some rock idol, or something. You could see Payton thinking: _Killing people. Far out."_
Tommy Krey, car theft, had been on the other side. He was still out on bail; Krey's attorney was dragging his feet on the trial. "The car owner's gonna move to California, I hear. Tommy's lawyer's looking for a plea," the deputy said.
Burrell Thomas had been across the aisle, and pled to simple assault, paid a fine. He was gone.
"I know Tommy, but I don't know the other two," Lucas said. Out of touch.
"Payton's from St. Paul, Rice Street. Basically a doper, sells real estate when he's straight," Sloan said. "I don't know Thomas either."
"Burrell's a head case," the deputy said. "They call him Rayon. Y'all know Becky Ann, the cardplayer with the huge hooters, see her down on Lake sometimes?"
"Sure." Lucas nodded.
"She was going with this super-tall black dude . . . ."
"Manny," said Sloan, and Lucas added, "Manfred Johnson."
"Yeah, that's him—he's a friend of Burrell's. Like from high school and maybe even when they were kids . . ."
• • •
"How's New York?" Sloan asked. They were in Sloan's unmarked car, poking into the south side of Minneapolis.
"Hot. Like Alabama."
"Mmm. I never been there. I mean New York. I understand it's a dump."
"It's different," Lucas said, watching the beat-up houses slide by. Kids on bikes, rolling through the summer. They'd called Krey's attorney, a guy who worked out of a neighborhood storefront. He could have Krey there in a half-hour, he said.
"How different? I mean, like, Fort Apache?"
"Nah, not that," Lucas said. "The main thing is, there's an infinite number of assholes. You never know where the shit is coming from. You can't get an edge on anything. You can't know about the place. Here, if somebody hijacks a goddamn Best Buy truck and takes off fifty Sonys, we got an idea where they're going. Out there . . . Shit, you could make a list of suspects longer than your dick, and that'd only be the guys that you personally _know_ might handle it. And then there are probably a hundred times that many guys that you don't know. I mean, a list longer than _my_ dick."
"We're talking long lists here," Sloan said.
"It's strange," said Lucas. "It's like being up at the top of the IDS Building and looking out a window where you can't see the ground. You get disoriented and you feel like you're falling."
"How 'bout that Bekker, though?" Sloan said enthusiastically. "He's a fuckin' star, and we knew him back when."
• • •
Tommy Krey was sitting on a wooden chair in his attorney's office. His attorney wore a yellow-brown double-knit suit and a heavily waxed hairdo the precise shade of the suit. He shook hands with Sloan and Lucas; his hands were damp, and Lucas smothered a grin when he saw Sloan surreptitiously wipe his hand on his pant leg.
"What can Tommy do for yuz?" the lawyer asked, folding his hands on his desk, trying to look bright and businesslike. Krey looked half bored, skeptical, picked his teeth.
"He can tell us what he and Michael Bekker talked about in jail," Lucas said.
"What are the chances of knocking down this car-theft . . ."
"You're gonna have to do that on your own," Lucas said, looking from the lawyer to Krey and back again. "Maybe Sloan goes in and tells the judge you helped on a big case, but there's no guarantees."
The lawyer looked at Krey and lifted his eyebrows. "What d'you think?"
"Yeah, fuck, I don't care," Krey said. He flipped his toothpick at the basket, rimmed it out, and it fell on the carpet. The lawyer frowned at it. "We talked about every fuckin' thing," Krey said. "And I'll tell you what: I been beatin' my brains out ever since he went out to New York, trying to figure out if he gave me, like, any _clues._ And he didn't. All we did was bullshit."
"Nothing about friends in New York, about disguises . . . ?"
"Naw, nothing. I mean, if I knew something, I'd a been downtown trying to deal. I know that his buddy, the guy who did the other kills, was an actor . . . so maybe it is disguises."
"What was he like in there? I mean, was he freaked out . . . ?"
"He cried all the time. He couldn't live without his shit, you know? It hurt him. I thought it was bullshit when I first went in, but it wasn't bullshit. He used to cry for hours, sometimes. He's totally fuckin' nuts, man."
"How about this Clyde Payton? He was in for some kind of dope deal, he was around Bekker."
"Yeah, he came in the day before I made bail. I don't know; I think he was a wacko like Bekker. Square, but wacko, you know? Kind of scary. He was some kind of businessman, and he gets onto the dope. The next thing he knows, he's busting into drugstores trying to steal prescription shit. He mostly sat around and cursed people out while I was there, but sometimes he'd get like a stone. He figured he was going to Stillwater."
"He did," said Sloan.
"Dumb fuck," said Krey.
"How about Burrell Thomas?"
"Now, there's something," Krey said, brightening. "Bekker and Burrell talked a lot. Rayon's one smart nigger."
Burrell's address was a vacant house, the doors pulled down, the floor littered with Zip-Loc plastic bags. They crunched across broken glass up an open stairway, found a burned mattress in one room, nothing in the other, and a bathtub that'd been used as a toilet. Flies swarmed in an open window as Sloan reeled back from the bathroom door.
"We gotta find Manny Johnson," Sloan said.
"He used to work at Dos Auto Glass," Lucas said. "Not a bad guy. I don't think he's got a sheet, but that woman of his . . ."
"Yeah." Manny's girlfriend called herself Rock Hudson. "She took twenty-five grand out of a high-stakes game down at the Loin last month. That's going around."
"She's a piece of work," Lucas agreed.
They found both Manny and Rock at the auto glass. The woman was sitting in a plastic chair with a box full of scratch-off lottery tickets, scratching off the silver with a jackknife blade, dropping the bad ones on the floor.
"Cops," she said, barely looking up when they came in.
"How are you?" Lucas asked. "Doing any good?"
"What d'ya want?"
"We need to talk to Manny," Lucas said. She started to heave herself to her feet, but Lucas put a hand in front of her head. "Go ahead with the tickets. We can get him."
Sloan had moved to the door between the waiting area and the workroom. "He's here," he said to Lucas.
They went back together. Johnson saw them, picked up a rag, wiped his hands. He was at least seven feet tall, Lucas thought. "Manny? We need to talk to you about Burrell Thomas."
"What's he done?" Johnson's voice was deep and roiled, like oil drums rolling off a truck.
"Nothing, far as we know. But he was bunked down at the jail next to Michael Bekker, the nut case."
"Yeah, Rayon told me," the tall man said.
"You know where we can reach him?"
"No, I don't know where he's living, but I could probably find him, tonight, if I walked around the neighborhood for a while. He usually goes down to Hennepin after nine."
"Bekker's chopping people up," Sloan said. "I mean chopping them up. I don't know if Burrell's got trouble with the cops, but if there's any way he could help us . . ."
"What?".
Sloan shrugged, picked up a can of WD-40, turned it in his hand, and shrugged. "We might be able to take a little pressure off, if he has another run-in with the cops. Or if your friend out there, if she . . ."
Johnson looked them over for a minute, then said, "You got a phone number?"
"Yeah," Sloan said. He fished a card out of his pocket. "Call me there."
"Like tonight," Lucas said. "This guy Bekker . . ."
"Yeah, I know," Johnson said. He slipped Sloan's card in his shirt pocket. "I'll call you, one way or another."
The drive to Stillwater cut another hour out of the day; the interview took ten minutes. Payton looked like an ex-college lineman, square, running to fat. He wasn't interested in talking. "What the fuck'd the cops ever do for me? I'm a sick man, and here I am in this cage. You guys can fuck yourselves."
They left him talking to himself, muttering curses at the floor.
"How're you gonna threaten him? Tell him you're gonna put him in jail?" Sloan asked as they walked back through the parking lot.
Lucas glanced back at the penitentiary. It looked like an old Catholic high school, he decided, inside and out, until you heard the steel doors open and shut. Then you knew it couldn't be anything but the joint . . . .
Johnson called Sloan's number a little after six o'clock. Burrell would talk and he'd meet Lucas at Penn's Bar, on Hennepin. Johnson would come down, to introduce them.
"Um, I got some shit to do at home," Sloan said.
"Hey, take off," Lucas said. "And thanks."
They shook hands, and Sloan said, "Don't take no wooden women."
Penn's bar had a sagging wooden floor and a thin mustachioed bartender who poured drinks, washed glasses, ran the cash register and kept one eye on the door. A solitary black hooker leaned on the bar, smoking a cigarette and reading a comic book, ignoring a half-drunk, pale-green daiquiri. The hooker picked up Lucas' eyes for a second, saw something she didn't like, and went back to her comic.
Farther toward the back, four men and two women stood around a coin-op pool table. Layers of cigarette smoke floated around them like the ghosts of autumn leaves. Lucas walked past the bar to the back, past the pool table, past a beat-up pay phone hung in an alcove next to a cigarette machine. He looked in the men's john, came back, walked around the crowd at the pool table. The men wore jeans and vests, with big wallets chained to their belts, and looked at him sideways as he went through. Johnson wasn't there. Neither was anyone who might be Burrell.
"What can I do you for?" the bartender asked, drying his hands on a mustard-stained towel.
"Bottle of Leinie's," Lucas said.
The bartender fished it out of a cooler and dropped it wet on the bar: "Two bucks." And then, tipping his head toward the back, "Looking for someone?"
"Yeah." Lucas paid and sat on a stool. The back-bar mirror ended before it got that far down, and Lucas stared into the fake walnut paneling opposite his stool, hitting on the beer, trying to straighten his schedule out.
If he didn't find Burrell quick, he'd have to stay over a day. Then he'd miss the early flight to Atlanta. Instead of getting into Charleston in the morning, he wouldn't make it until the afternoon and probably wouldn't get out until the next day. Then he'd have to think of an excuse for the New York people.
The hooker rapped on the bar with her knuckles, nodded at the daiquiri, got a new one. She wore a pale-green party dress, almost the color of the drink. She caught his eyes again, let her gaze linger this time. Lucas didn't remember her. He'd known most of the regulars when he was working, but he'd been off the streets for months now. A week is forever, on the streets. A whole new class of thirteen-year-old girls would be giving doorway blow jobs to suburban insurance agents who would later be described in court documents as good fathers . . . .
Lucas was halfway through the beer when Johnson walked in, out of breath, as though he'd been running.
"Jesus, Davenport," he said. "Missed the bus." He looked down the bar at the hooker as Lucas swiveled on the stool.
"Where is he?" Lucas said.
Johnson's face lit up. "What'd you mean, where is he? He's right there."
Lucas looked past the hooker to the back of the bar; all the pool players were white.
"Where?"
Johnson started to laugh, lifted a leg and slapped a thigh. "You sittin' next to him, man."
The hooker looked at Lucas and said, in a voice an octave too low, "Hi, there."
Lucas looked at the hooker for a second, rereading the features, and closed his eyes. Transvestite. In a half-second, it all fell into place. Goddamn Bekker. This was how he got close to the women and the tourist males. As a woman. With the right makeup, at night, with his small, narrow-shouldered body. That was how he got out of the New School . . . .
God damn it.
"Did you tell Bekker how to . . . do this?" Lucas asked, gesturing at the dress. "The dress, the makeup."
"We talked about it," Thomas said. "But he was a sick motherfucker and I didn't like talking to him."
"But when you talked about it . . . was he real interested, or did you just talk?"
Thomas tipped his head back, looked up at the ceiling, remembering. "Well . . . he tried it. A couple of things." He hopped off the bar stool and walked away from Lucas and Johnson, moving his hips, turned and posed. "It ain't that easy to get just the right walk. If you forget halfway through the block, it ruins your whole image."
The bartender, watching, said, "Are you guys gay?"
"Cop," said Lucas. "This is official."
"Forget I asked . . ."
"I won't forget, honey," Thomas said, licking his lower lip.
"You fuckin' . . ."
"Shut up," Lucas snapped, poking a finger at the bartender. He looked back at Thomas. "But did he do it? The walk?"
"Couple times, a few times, I guess. You know, we _did_ talk about it, when I think back. Not so much about how good it feels, but how to do it. You know, gettin' the prosthetic bras and like that. He'd make a good-lookin' girl, too, 'cept for the scars."
"You think so?" Lucas asked. "Is that a professional opinion?"
"Don't dick me around, man," Thomas said, flaring.
"I'm not. That's a real question. Would he make a good woman?"
Thomas stared at him for a minute, decided the question was real: "Yeah, he would. He'd be real good at it. 'Cept for the scars."
Lucas hopped off the bar stool, said thanks, and nodded to Johnson: "We owe you. You need something, talk to Sloan."
"That's all?" asked Thomas.
"That's all," Lucas said.
Lucas called Fell from the pay phone at the back of the bar. When she answered, he could hear the television going in the background, a baseball game. "Can you get to Kennett? Right now?"
"Sure."
"Tell him we've figured out how Bekker is doing it," Lucas said. "How he's staying out of sight on the streets, getting out of the New School."
"We have?"
"Yeah. I just talked to his former next-door neighbor at the Hennepin County Jail, name of Rayon Thomas. Nice-looking guy. Good makeup. Great legs. He's wearing a daiquiri-green party dress. He gave Bekker lessons . . . ."
After a moment of silence, she breathed, "Sonofabitch, Bekker's a woman. We're so fuckin' stupid."
"Call Kennett," Lucas said.
"You haven't talked to anyone?" she asked.
"I thought you'd like to break it."
"Thanks, man," Fell said. "I . . . thanks."
## CHAPTER
20
Bekker could count the drops, each and every one, as the shower played off his body. The ecstasy did that: two tiny pills. Gave him the power to imagine and count, to multiply outrageous feelings by ineffable emotions and come up with numbers . . . .
He turned in the shower, letting jets of water burn into him. He no longer used the cold water at all, and the stall was choked with heat and steam, his body turning cherry red as the old skin scalded away. And as he turned, his eyes closed, his head tipped back, his hands beneath his chin, his elbows close together, on his belly, he could count all the drops, each and every one . . . .
He stayed in the shower until the water ran cold, then, shivering, blue, annoyed, he leaped out. What time was it? He walked to the end of the room where he'd fitted a black plastic garbage bag over a barred basement window, and peeled back a corner of the plastic. Dark. Midnight. That was good. He needed the night.
Bekker walked back toward the bed, felt the stickiness on the soles of his feet and looked down. He needed to wash the floor. The sight of the dried blood on the floor reminded him of the cut. He looked at his arm, rolled it between his thumb and forefinger. The cut was painful, but the ants were gone.
He caught sight of himself in a wall mirror, his furrowed face. He went into the bathroom and washed his face, grimacing at the sight of the scars. They were in long jagged rows, raised above the soft skin around them. The gunsight cuts had been sewn closed by an emergency-room butcher, instead of a qualified plastic surgeon.
He thought of Davenport, Davenport's teeth, the eye-teeth showing, his eyes, the gun swinging, battering . . .
He sighed, came back, shaken, staring at his face in the mirror. He put the makeup on mechanically, but carefully. Cover Mark to hide the scars, then straight, civilian makeup. Max Factor New Definition. Cover Girl nail polish. Suave styling spritz, to pull his blond hair down to cover his jawline, which was a bit too masculine.
The lipstick was last. Lipstick the color of a prairie rose. Just a touch. He didn't want to be mistaken for a harlot . . . . He made kisses at the mirror, smoothed the lipstick with his tongue, blotted it with toilet paper. Just right.
Satisfied, finally, he went to the chest, picked out underwear, got the prosthetic bra and sat on the bed. He'd shaved his legs the night before, and they were just getting prickly. Bekker was fair-haired, fine-haired: even if he hadn't shaved, his legs wouldn't have been a problem. But he did shave, to capture the feel. Rayon had said that was important, and Bekker understood—or he'd understood at the time. You had to live the part, feel the part. He flashed. A woman hurrying behind him, afraid of the dark parking ramp. Live the part . . .
The panty hose slid smoothly up his leg; he'd discovered the technique of gathering them, slipping them up piece by bit-bit-bit. When the hose were on, he stood and looked at himself in the dressing mirror; he looked like a fencer, he thought, bare chest and tights. He posed, turning sideways. A little full in the front. He reached into the panty hose and arranged his penis, pushing it down and under, tight, pulling the hose up to hold it in place. Posed again. Good.
The bra was next. He disliked it: it was cold and awkward, and cut into the muscles of his shoulders. But it gave him the right look and even the right feel. He snapped it in back, and again checked the mirror. With his soft blond hair, falling naturally now to his shoulders—no more wigs—he _was_ a woman. Whitechurch had certainly been convinced. Bekker flashed: the look on Whitechurch's face as the _realization_ came to him, and the gun came up . . .
He picked out a medium-blue blouse with a high collar and the remnants of shoulder padding, a conservative, midcalf-length pleated skirt, and dark gym shoes with thick walking soles. With the breast prosthetics and his narrow shoulders, he had the figure of a woman, but his hands and feet might yet give him away.
They were simply too big, too square: he wore size ten men's shoes. But when he wore dark women's gym shoes, the size was not so obvious. As a woman he was taller than average, but not awkwardly so. And people expected blondes to be tall. Hiding his hands was a bigger problem . . . .
When he'd finished dressing, he looked in the mirror. Fine. Excellent. The big shoulder bag was something he might keep dressier shoes in, wearing the gym shoes to walk back and forth to the parking ramp. Yuppie. He added a necklace of synthetic pearls, picked up a bottle of Poison by Christian Dior, dabbed it along his throat, on the inside of his wrists. The perfume was too flowery, and he deliberately used too much. Perfume, Rayon told him, was a feminine, psychological thing. The odor of perfume alone might subliminally convince, in close quarters . . . .
There. Ready. He touched himself at the pit of his throat, and remembered that he'd seen his late wife do that, touch herself there, a sort of completion. He stepped to the mirror again, to take in the whole ensemble, and spontaneously laughed with the joy of it.
Beauty was back.
Beauty stepped carefully through the weeds to the lean-to garage, careful not to snag the hose. He left the car lights out, drove it to the gate, looked up and down the street, unlocked the gate, drove through, relocked it behind himself. He sat in the car for a moment, trying to think.
The parking garage at Bellevue was locked in his brain. Bellevue. He reached across the floor to his purse, found the bag, shook out a greenie: PCP. Popped one, two. Folded the bag and dropped it back in the purse, turned left. Careful. Bellevue? The hands on the steering wheel took him there, rolling through the dimly lit streets, precisely, evenly. A woman? Yes. Women were smaller and handled more easily, after they'd been taken. He recalled the struggle with Cortese, wedging the deadweight into the backseat of the Bug.
And women, he thought with sudden clarity and some curiosity, lasted longer . . . .
The guard nodded. He recognized the attractive blonde in the old Volkswagen Bug. She'd been there before . . . .
Bekker took the car to the top floor, which was virtually deserted. A red Volvo sat in a corner and looked like it might have been there for a couple of days. Two other cars were widely spaced. The garage was silent. He got his bag from the passenger-side floor, with the tank of anesthetic and the stun gun.
Bekker flashed: Cortese, the first one. Bekker'd hit him with the stun gun, had ridden him like a . . . No image came for a moment, then a hog. A heavy, midwestern boar, a mean brute. Bekker had ridden him down in the alley behind the Plaza, then used the mask. The power . . .
A car door slammed somewhere else in the garage; a hollow, booming sound. An engine started. Bekker went to the elevator, pushed the down button, waited. A sign on the wall said: "REMOVE VALUABLES FROM CAR: Although this ramp is patrolled, even locked cars are easily entered. Remove all valuables."
The first hit of PCP was coming on, controlling, toughening him, giving his brain the edge of craft it needed. He glanced around. No camera. He walked slowly down the stairs past the cashier, around the corner toward the main entrance of the hospital. The sidewalk that led to the entrance was actually built as a ramp, slanting down between the parking ramp and a small hospital park. Bekker walked down the ramp, paused, then went left into the park, sat at a bench under a light.
Outside, the night was warm and humid, the smell of dirty rain and cooling bubble gum. A couple on the street were walking away from him, the man wearing a straw hat; the hat looked like an angel's halo at that distance, a golden-white oval encircling his head.
Then: A main hospital door opened and a woman walked out. Headed toward the ramp, digging in a purse for keys. Bekker got up, started after her. She paused, still digging. Bekker closed. The woman was big, he realized. As he got closer, he saw she was too big. A hundred and eighty or two hundred pounds, he thought. Moving her would be difficult.
He stopped, turned, lifted a foot so he could look at the sole of his left shoe. Watch women, Rayon had told him. Watch what they do. Bekker had seen this, the stop, the check, the look of anger or disgust, depending on whether a heel was broken or she'd simply stepped in something, and then a turn . . . .
He turned, as though he might be going somewhere to fix whatever he was looking at, walked away from the heavy woman, back down into the park. He might be waiting for someone inside, might even be grieving. There were cops around, nobody would bother him . . . .
Shelley Carson was a graduate nurse. She ran an operating suite, took no crap from anyone.
And she was just the right size.
_Bite-size,_ Bekker's brain said when he saw her.
At five-two, she barely reached a hundred pounds when she was fully dressed. Aware of her inviting size, she was careful about the ramp. Tonight she walked out with Michaela Clemson, tall, rangy, blonde and tough; a lifelong tennis player, both a nurse and a surgical tech. They were still in uniform, tired from the day.
"Then you heard what he said? He said, 'Pick it up and put it where I told you to in the first place,' like I was some kind of child. I am definitely going to complain . . ." Clemson was saying.
Bite-sized Shelley Carson encouraged her: nurses were not less than doctors, they were members of a different profession. They should take no shit. "I'd certainly go in . . . ."
"I just can't ignore it this time," the blonde said, building her courage. "The asshole is a bad surgeon, and if he'd spend more time working on his surgery and less time trying to pull rank . . ."
Bekker slid in behind them. They saw him, peripherally, but neither really looked at him until they started into the ramp together, and then up the stairs.
"I definitely would," the small one was saying. Her dark hair was cut close to the head, like a helmet, with little elfin points over her ears.
"Tomorrow afternoon at three o'clock I'm going to march in . . ." The blonde looked down at Bekker, then back up at her friend. Bekker climbed behind them, one hand on the stun gun.
Halfway up, the blonde said, "Tomorrow, I go for it."
"Do it," said the elf. "See you tomorrow."
The blonde broke away, stepping into the main part of the ramp, peering out. "All clear," she said. The blonde started toward a Toyota. Bekker and the dark-haired woman continued up, Bekker's heels rapping on the stairs.
"We have an arrangement," the elf said, looking down at Bekker. "If one of us has to go out alone, we watch each other."
"Good idea," Bekker squeaked. The voice was the hard part. Rayon had said it would be. Bekker put a hand to his mouth and faked a cough, as though his voice might be roughed by a cold, rather than forty years of testosterone.
"This parking garage, somebody's going to get attacked here someday," the woman said. "It really isn't safe . . . ."
Bekker nodded and went back to the purse. The elf looked at him, a puzzled look, something not quite right. But what? She turned away. Turn away from trouble. Bekker followed her out at the top floor, heard the Toyota's engine start below. Brought the stun gun out, got the tank ready in the bag. Heard the hiss. Felt the action in his feet . . .
The woman saw him moving. A fraction of a second before he was on her, she took in the violence of his motion and started to turn, her eyes widening in reflex.
Then he had her. One hand over her mouth, the other pressing the stun gun against her neck. She went down, trying to scream, and he rode her, pressing the stun gun home, holding it . . . .
She flapped her arms like the wings of a tethered bird. He dropped the stun gun, groped for the tank, found it, flipped the valve and clapped the mask over her face. He had her now, his hair a bush around his head, his eyes wide, feral, like a jackal over a rabbit, breathing hard, mouth open, saliva gleaming on his teeth.
He heard the sound of the Toyota going down the ramp as the bite-sized woman's struggles weakened and finally stopped. He stood up, listening. Nothing. Then a voice, far away. The little woman was curled at his feet. So sweet, the power . . .
Bekker worked all night. Preparing the specimen—wiring the gag, immobilizing her. Taking her eyelids; he held them in the palms of his hands, marveling; they were so . . . interesting. Fragile. He carried them to a metal tray, where he'd collected some others. The others were drying now, but kept their form, the lashes still shiny and strong . . .
Shelley Carson died just before seven o'clock, as silently as all the rest, the gag wired around her skull, her eyes permanently open. Bekker had crouched over her with the camera as she died, shooting straight into her eyes.
And now he sat in his stainless-steel chair and gazed at the proof of his passion, eight ultraviolet photos that clearly showed something—a radiance, a presence—flowing from Carson as she died. No question, he exulted. No question at all.
_Dink._
The intercom bell. It cut through the sense of jubilation, brought him down. Old bitch. Mrs. Lacey got up early, but habitually slumped in front of the television until noon, watching her morning shows.
_Dink._
He went to the intercom: "Yes?"
"Come quick," she squawked. "You have to see, you're on the television."
_What?_ Bekker stared at the intercom, then went quickly to the bed, picked up his robe, wrapped himself, put fluffy slippers on his feet. The old lady didn't see very well, didn't hear very well, he could pass . . . and he still had on his makeup. On television? As he passed the dresser he slipped two tabs off the tray, popped them, as brighteners. What could she mean?
The first floor was dark, musty, a thin orangish morning light filtering through the parchmentlike window shades. The second floor was worse, the odor of marijuana hanging in the curtains, a stench of decaying cat shit, the smell of old vegetables and carpet mold. And it was dark, except for the phosphorescent glow of the tube.
Mrs. Lacey was standing, staring at the television, a remote control in her hand. Bekker was there on the screen, all right. One of the photos that had plagued him, had kept him off the street. But in this photo, he was a woman and a blonde. The details were perfect:
" . . . _credited to Detective Barbara Fell and former Minneapolis Detective Lieutenant Lucas Davenport, who had been brought to New York as a consultant_ . . ."
Davenport. Bekker was struck by a sudden dizziness, a wave of nausea. Davenport was coming; Davenport would kill him.
"But . . ." said Mrs. Lacey, looking from the screen to Bekker.
Bekker steadied himself, nodded. "That's right, it is me," he said. He sighed. He hadn't expected the old woman to last this long. He stepped carefully across the carpet to her.
She turned and tried to run, a shuffling struggle against age and infirmity, gargling in terror. Bekker giggled, and the cats, hissing, bounded across the overstuffed furniture to the highest shelves. Bekker caught the old woman at the edge of the parlor. He put the heel of his left hand against the back of her skull, the cup of his right under her chin.
"But . . ." she said again.
A quick snap. Her spine was like a stick of rotten wood, cracked, and she collapsed. Bekker stared down at her, swaying, the brightener tab coming on.
"It is me," he said again.
## CHAPTER
21
Most visitors came through O'Dell's office; when the knock came at Lily's unmarked office door, she looked over the top of her _Wall Street Journal_ and frowned.
There was another light knock and she took off her half-moon reading glasses—she hadn't let anyone see them yet—and said, "Yes?"
Kennett stuck his head in. "Got a minute?"
"What're you doing down here?" she asked, folding the paper and putting it aside.
"Talking to you," he said. He stepped inside the door, peeked through a half-open side door into O'Dell's office, and saw an empty desk.
"He's at staff," Lily said. "What's going on?"
"We've papered the town with the female Bekker picture," Kennett said, dropping into her visitor's chair. Small talk. He tried a smile, but it didn't work. "You know Lucas got it, the cross-dressing thing. It wasn't Fell."
"I thought maybe he did," Lily said. "He wants Fell to do well."
"Nice," he said, his voice trailing off. He was looking at her as though he were trying to see inside her head.
"Let's have it," she said finally.
"All right," he said. "What do you know about this Robin Hood shit that O'Dell is peddling?"
Lily was surprised—and a small voice at the back of her head said that was good, that look of surprise. "What? What's he peddling?"
Kennett looked at her, eyes blinking skeptically, as though he were reevaluating something. Then he said, "He's been putting out shit about Robin Hood, the so-called vigilantes. I've got the feeling that the fickle finger is pointed at my ass."
"Well, Jesus," Lily said.
"Exactly. There aren't any vigilantes. It's all bullshit, this Robin Hood business. But that doesn't mean he can't fuck me up. If they think they've got a problem . . ." He pointed a thumb at the ceiling, meaning the people upstairs, "And they can't find anybody, they might just want to hang somebody anyway, to cover their asses."
"Boy . . ." Lily shook her head. "I've got a pretty good line on what O'Dell's doing, but I don't know anything like that. And I'm not holding out on you, Richard. I'm really not."
"And I'm telling you, he's behind it."
Lily leaned forward. "Give me a few days. I'll find out. Let me ask some questions. If he's doing it, I'll tell you."
"You will?"
"Of course I will."
"All right." He grinned at her. "It's, like, when you're a lieutenant and down, you've got friends and lovers. When you're a captain or above, you've got allies. You're my first ally-lover."
She didn't smile back. She said: "Richard."
The smile died on his face. "Mmm?"
"Before I risk my ass—you're not Robin Hood?"
"No."
"Swear it," she said, looking into his eyes.
"I swear it," he said, without flinching, looking straight back at her. "I don't believe there is such a guy. Robin Hood is a goddamn computer artifact."
"How?"
He shrugged. "Flip a nickel five hundred times. The events are random, but you'll find patterns. Flip it another five hundred times, you'll still find patterns. Different ones. But the pattern doesn't mean anything. Same thing with these computer searches—you can always find patterns if you look at enough numbers. But the pattern's in your head; it's not real. Robin Hood is a figment of O'Dell's little tiny imagination."
Her eyes narrowed: "How'd you find out so much about what he's doing?"
"Hey, I'm in intelligence," he said, mildly insulted by the question. "The word gets around. I thought his little game was pretty harmless until my name started popping up."
She thought about it a minute, then nodded. "All right. Let me do some sneaking around."
## CHAPTER
22
Lucas called Darius Pike in Charleston and gave him the plane's arrival time, then met Sloan and Del downtown. They hit a sports bar, talking, remembering. Lucas was long out of the departmental gossip—who was kissing whose ass, who was shagging who. Sloan went home at one o'clock and Lucas and Del wound up in an all-night diner on West Seventh in St. Paul.
" . . . shit, I said, gettin' married was okay," Del said. "But then she started talking about a kid. She's, like, forty."
"Ain't the end of the world," Lucas said.
"Do I look like _Life with Father?_ " Del asked. He spread his arms: he was wearing a jeans jacket with a black sleeveless tank top. An orange and black insignia on the sleeve of the jacket said, " _Harley-Davidson_ —Live to Ride, Ride to Live." He had a five-day beard, but his eyes were as relaxed and clear as Lucas had ever seen them.
"You're looking pretty good, actually," Lucas said. "A year ago, man, you were ready for the junk heap."
"Yeah, yeah . . ."
"So why not have a kid?"
"Jesus." Del looked out the window. "I kinda been asking myself that."
Del peeled off at three o'clock and Lucas went home, opened all the windows in the house, and began writing checks to cover the bills that had arrived with the mail. At five, finished with the bills, and tired, he closed and locked all the windows, went back to the bedroom and repacked his overnight bag. He called a cab, had the driver stop at a SuperAmerica all-night store, bought two jelly doughnuts and a cup of coffee, and rode out to the airport.
The plane taxied away from the terminal at six-thirty. The stewardess asked if he wanted juice and eggs.
"I'm gonna try to go to sleep," he said. "Please, please don't wake me up . . . ."
The fear got him as the takeoff run began, the sense of helplessness, the lack of control. He closed his eyes, fists clenched. Got off the ground with body English. Held his breath until the engine noise changed and the climb rate slowed. Cranked back the seat. Tried to sleep. A while later, he didn't know how long, he realized that his mouth tasted like chicken feathers, and his neck hurt. The stewardess was shaking his shoulder: "Could you bring your seat upright, please?"
He opened his eyes, disoriented. "I was sleeping," he groaned.
"Yes," she said in her most neutral voice. "But we're approaching Atlanta, and your seat . . ."
"Atlanta?" He couldn't believe it. He never slept on airplanes. The plane's left wing dipped, and they turned on it, and, looking down, he could see the city of Atlanta, like a gritty gray rug. Ten minutes later, they were down.
The Atlanta airport was straight from _RoboCop,_ with feminine machine voices issuing a variety of warnings just below the level of consciousness, and steel escalators dropping into sterile tile hallways. He was glad to get out, though the flight to Charleston was bad. He fought the fear and managed to compose himself by the time the plane was on the ground.
Pike was waiting inside the small terminal, a stolid black man wearing a green cotton jacket over a white shirt and khaki pants. When his jacket moved, Lucas could see a half-dozen ballpoint pens clipped to his shirt pocket and a small revolver on his belt.
"Lucas Davenport," Lucas said, shaking hands.
"I gotta car," Pike said, leading the way. "How's New York?"
"Hotter'n here," Lucas said.
"This is nothin'," Pike said. "You ought to be here in August."
"That's what they say in New York . . . ."
They left the airport at speed. Lucas, disoriented, asked, "Where's the ocean?"
"Straight ahead, but the city's not really on the ocean. It's kind of like . . . Manhattan, actually," Pike said. "There's a river coming in on both sides, and they meet, and that's the harbor, and then you gotta go on out past the Fort to get into the ocean."
"Fort Sumter?"
"That's it," Pike said.
"I'd like to see it sometime. I've been going to battlefields. Tell me about Reed."
Pike whipped past a gray Maxima, took an off-ramp, then turned left at the bottom. The street was cracked, the borders overgrown with weeds and scrub. "Reed is a stupid motherfucker," he said matter-of-factly. "I get mad talking about it. His old man has lived here all his life, runs a garage and gas station, does the best body work in town, and makes a ton of money. And Red did good in high school. Did good on his tests and got into Columbia University on a scholarship. The silly fuck goes up to New York and starts putting junk up his nose, the cocaine. Hanging out in Harlem, coming back here and talking shit. Then he didn't come back anymore. The word was, he was putting it up his nose full-time."
"Huh. How long's he been back?"
"Few weeks," Pike said. "I feel bad for his folks."
"Is he staying?"
"I don't know. When he first got back, there were a couple of rumbles from Narcotics that he was hanging out with the wrong people. But I haven't heard that lately. Maybe something changed."
Lucas hadn't thought about what Charleston might look like, but as they drove through, he decided it was just right: Old South. Clapboard houses with peeling paint, and weird trees; bushes with plants that had leaves like leather, and spikes. A few palms. A lot of dirt. Hot.
The Reed garage was a gray concrete-block building sitting side by side with a Mobil gas station and convenience store. All but one set of the gas pumps had a car parked next to them, and uniformed attendants moved around cleaning windshields and checking oil. "You come in here, they wipe your windshield, check your oil, put air in your tires. The only place you'll find it," Pike said. "That's why Don Reed makes the money he does."
He killed the engine in the body shop's parking lot and Lucas followed him into the shop office. The office smelled of motor oil, but was neatly kept, with plastic customer chairs facing a round table stacked with magazines. Behind a counter, a large man was hunched over a yellow-screen computer, poking at a keyboard one finger at a time. He looked up when they came in and said, "Hey, Darius."
"Hey, Don. Is Red around?"
Reed straightened up, his smile slipping off his face. "He done somethin'?"
Pike shook his head and Lucas said, "No. I'm from New York. Your son witnessed a shooting. He was a passerby. I just need to talk to him for a couple of minutes."
"You sure?" Reed asked, a hostile tone scratching through. "I got a lawyer . . ."
"Look: You don't know me, so . . . But I'm telling you, with a witness standing here, that all I want to do is talk. There's no warrant, no anything. He's not a suspect."
Reed regarded Lucas coolly, then finally nodded. "All right, come on. He's out back."
Red Reed was coming out of a paint room when they found him, a plastic mask and hat covering his head. When he saw his father and the two cops, he pulled off the protective gear and waited uncertainly by the paint room door. He was tall, too thin, with prominent white teeth.
"Police to talk to you. One from New York," his father said. "I'm gonna listen." Red Reed looked apprehensive, but nodded.
"Can we find a place to sit?" Lucas asked.
The elder Reed nodded: "Nobody in the waiting room . . . ."
Lucas took Bobby Rich's report from his pocket, unfolded it, and led Red Reed through it, confirming it bit by bit.
"White-haired guy," Lucas said. "Thin, fat?"
"Yeah. Skinny, like."
"Dark? Pale? What?"
"Tan. He was, like, tan."
"What was the scene like, when Fred Waites was shot?"
"Well, man, I wasn't right there. I saw the car go by and I thought I saw a gun and I headed the other way. I heard the shooting, saw the car."
"What kind of car?"
"I don't know, man, I wasn't paying attention to that," Reed said. He was looking at his hands. Pike moved impatiently, and Reed's father looked out the door but didn't say anything. Reed's eyes wandered to his father, then back to Lucas.
"What time was it?" Lucas asked.
"I didn't have a watch . . . ."
"I mean, afternoon, evening, night?"
Reed nervously licked his lips, then seemed to pick one: "Evening."
"It was three o'clock in the afternoon, Red," Lucas said. "Bright daylight."
"Man, I was fucked up . . ."
"You don't know what kind of car it was, but you could see inside that the guy was white-haired, skinny and tanned? But you didn't see anything about the other guys? Red . . ." Lucas glanced at Don Reed. "Red, you're lying to us. This is an important case. We think the same guys shot a cop and, before that, a lawyer."
"I don't know nothing about that," Reed said, now avoiding everyone's eyes.
"Okay, I don't think you do. But you're lying to me . . ."
"I'm not lying," Reed said.
Don Reed turned to face his son and in a harsh, cutting voice said, "You remember what I told you? No bullshit, no lies, no dope, no stealing, and we'll try to keep you alive. And you're lying, boy. There never was a time, from when you were a little baby, that you didn't know what kind of car was what—and you see a man and know he's got white hair and a tan, and you don't know what car he was in? Horseshit. You're lying. You stop, now."
Lucas said, "I want to know how much John O'Dell had to do with it."
Reed had been staring miserably at his feet, but now his head popped up.
"You know Mr. O'Dell?"
"Aw, shit," Lucas said. He stood up, walked once around the tiny room, whacked the spherical Lions Club gum machine with the palm of his hand, then pinched the bridge of his nose, closing his eyes. "You're fuckin' working for O'Dell."
"Man . . ." said Reed.
"O'Dell a dope pusher?" Don Reed asked, voice dark, angry.
"No," Lucas said. "He's about the fifth most important cop in New York."
The two Reeds exchanged glances, and Pike asked, "What's going on?"
"A goddamned game, pin the tail on the donkey," Lucas said. "And I'm the jackass."
He said to Reed, "So now I know. I need some detail. Where'd you meet him, how'd you get pulled in on this . . ."
Reed blurted it out. He'd met O'Dell at a Columbia seminar. O'Dell spoke three times, and each time, Reed talked to him after class. Harlem was different than an Irish cop could know, Reed said. The fat cop and skinny southerner argued about life on the streets; went with a few other students and the professor to a coffee shop, talked late. He saw O'Dell again, in the spring, but he was into the dope by then. Busted in a sweep of a crack house, called O'Dell. The arrest disappeared, but he was warned: never again. But there _was_ another time. He was arrested twice more for possession, went to court. Then a third time, and this time he had a little too much crack on him. The cops were talking about charging him as a dealer, and he called O'Dell. He got simple possession, and was out again.
Then O'Dell called. Did he know anybody, a crook, with a connection to a cop? To a detective? Well, yes . . .
"Sonofabitch. It was too neat, it had to be," Lucas said.
"What the fuck is going on?" Pike asked again.
"I don't know, man," Lucas said. To Reed, he said, "Don't call O'Dell. You're out of this and you want to stay out. Whatever's going on here, and it's pretty rough, doesn't have anything to do with you. You'd best lay low."
"He's out," Don Reed said, looking at his son.
Reed's head bobbed. "I don't want nothing more to do with New York."
On the way back to the airport, Pike said, "I don't think I'd like New York."
"It's got some low points," Lucas said. He took a card from his pocket diary, scribbled his home phone number on the back of it. "Listen, thanks for the help. If you ever need anything from New York or Minneapolis, call me."
The flight to Atlanta was bad, but on the way to New York, the fear seemed to slip away. Lucas had reached a tolerance level: his fifth flight in three days. He'd never flown that much in his life. More or less relaxed, he found a notepad in his overnight case and doodled on it, working it out.
Bobby Rich hadn't been assigned to work the case because he had the best qualifications—he'd been assigned simply because he knew a guy who knew Red Reed. So that Red Reed could call his friend and insist that the friend pass information to the cops about the shooting of Fred Waites.
Except that Reed hadn't been there at all. The man with white hair and the deep tan was an O'Dell invention. Lucas grinned despite himself. In a crooked way, it was very nice: lots of layers.
He closed his eyes, avoiding the next question: Did Lily know?
At La Guardia he saw a copy of the _Times_ with Bekker as a blond woman. He bought a copy, queued for a cab, got a buck-toothed driver who wanted to talk.
"Bekker, huh?" buck-tooth said, his eyes on the rearview mirror. He could see the picture on the front of the paper as Lucas read the copy inside. "There's a goofball for ya. Dressed up like a woman."
"Yeah."
"This last one, man, took her right out of a parking garage. Girlfriend says Bekker was right there with them, could've took them both."
Lucas folded the paper down and looked at the back of the driver's head. "There's another one? Today?"
"Yeah, this morning. They found her in a parking lot with the wire gag and the cut-off eyelids and the whole works. I say, when they get him, they ought to hang him off a street sign by his nuts. Be an example."
Lucas nodded and said, "Listen, forget about the hotel. Take me to Midtown South."
## CHAPTER
23
Carter, Huerta and James were huddled together over a tabloid newspaper in the coordinating office, all three of them with Styrofoam coffee cups in their hands. Lucas looked in and James said, "Kennett's down in the corner office, he wants to see you."
"Have you seen Barbara Fell?" Lucas asked.
"Gone home." There was a rapid-fire exchange of glances among the three cops, a vein of thin amusement. They knew he was sleeping with Fell.
"Anything happening?"
"About a thousand sightings on Bekker, including three good ones," Carter said. "He's driving a Volkswagen Bug . . . ."
"Jesus, that's terrific," Lucas said. "Who saw him? How'd you get the car?"
"Two witnesses last night at the parking ramp. The Carson woman's girlfriend and the cashier. The girlfriend is a sure thing—she even told us he was wearing too much Poison. That's a perfume . . ."
"Yeah."
" . . . And the cashier remembers the blond part, and says she—he—was driving an old Volkswagen. He remembers because it looked like it was in pretty good shape and he wondered if Bekker was an artist or something. He thinks it was dark green or dark blue. We're running it through the License Bureau right now, but the Volkswagen part isn't public yet. If he goes outside now, he's gonna have to go in a car. And we're stopping every Bug in Midtown."
"You said three people . . . ."
"The third's a maybe, but pretty definite. The night clerk in a bookstore down in the Village says he remembers the face very clearly, says it was Bekker. He says he was buying some weird book about torture."
"Huh."
"We're getting close," Carter said. "We'll have him in two or three days, at the outside."
"I hope," Lucas said. "Any returns on that stun-gun business?"
"Three. Nothing."
"Phones?"
"Nope. Goddamn rat's nest."
"Okay . . ."
Lucas started to turn away, and Carter said, "You've seen the papers?"
"With Bekker? Yeah . . ."
"No, that was this morning; the afternoon paper . . ." Huerta picked up the paper they'd been looking at, closed it, and handed it to Lucas. On the cover was a woman's face, eyes staring; before the headlines reached the brain, the terror of the face came through, then the words: _"Kill #8—Bekker Death Pix._ "
"This legit?" Lucas asked.
"That's Carson," Carter said grimly. "He sent notes and photos to three newspapers and two TV stations. They're using them."
"Jesus . . ."
From down the hall, he heard a woman's voice.
Lily.
He walked down to the corner, found the room in semidarkness, the door open. He knocked, standing back, and Kennett said, "Yeah?"
Lucas stuck his head in. "Davenport," he said.
"Come on in. We were just talking about you," Kennett said. He was sitting in a visitor's chair in front of a standard-issue metal desk, his feet up. His shirt collar was open, and his bright Polynesian Gauguin tie was draped across a stack of phone books at the front edge of the desk. Lily sat in another chair at the side of the desk, facing him.
"Fuckin' photographs," Lucas said.
"The shit is hitting the fan," Kennett said grimly. "First the New School thing and now the pictures. The mayor had the commissioner on the carpet. You could hear the screaming in Jersey."
Lucas dragged a third chair around, bumped Kennett. "Move your ass over so I can get my feet up."
"And me with a fuckin' bad heart," Kennett mumbled as he moved.
"You told Fell about the transvestite thing," Lily said. She pushed the phone books out of the way, picked up the necktie.
Lucas shrugged, sat down, put his feet up. "We talked it over and decided it was likely."
"That came at a good time. We told everybody that Carson'll probably be the last, that we've pretty much got him pinned down," she said.
"Should have thought of it sooner, the cross-dressing," Kennett said glumly. "The one before was a lesbian, we knew that. We should have seen that she wouldn't let a strange guy get too close, not outside a lesbian bar."
"Hell, you did everything right . . ." Lily began.
Kennett interrupted: "Everything but catch him . . ."
"He's pinned."
"We fuckin' hope," Kennett said.
Lily had been rolling the tie in her fingers, and now she looked down at the bare-breasted Polynesian woman, shook her head and said, "This is the craziest tie."
"Don't knocker it," said Kennett, then slapped his leg and laughed at the pun, while Lily rolled her eyes.
"You were jerking me around, Gauguin and Christian Dior," Lucas said to Kennett. He looked at Lily. "He told me this Gauguin dude was Christian Dior's necktie partner."
Lily laughed again, and Kennett said, "How do you know he wasn't?"
"Looked him up," Lucas said. "He died in 1903. He was associated with the symbolists."
"Now if you knew what a symbolist was, you'd be in fat city," Lily said.
"It was the use of color specifically for its symbolic impact, the emotional and intellectual impact," Lucas said. "Which makes sense. Some holding cells are painted bubble-gum pink for the same reason. The color cools people out."
Kennett, staring, said, "I never fuckin' thought of that."
"Carter tells me you'll have Bekker in three days at the outside," Lucas said.
"That fuckhead. That's the kind of talk that gets us in trouble," Kennett grumbled. "We'll get him soon, but I wouldn't bet on the three days. If he's got food and water, he could hole up."
"Still . . ."
"I figure no more than a week," Kennett said. "He'll break. I just hope I'm still working for the goddamn police department when it happens. I mean, people are _pissed._ These fuckin' pictures, man: the mess at the New School was nothing, compared to this."
"People think cops . . ." Lucas started.
But Lily was shaking her head. "It's not the people, it's the politicians. People understand you can't always catch a guy immediately; most of them do, anyway. But the politicians think _they've_ got to do something, so what they do is run around and scream and threaten to fire people."
"Mmmm. A week," Lucas said. "That's a long time, in ward-heeler years."
"Anxious to get home?" Kennett asked.
"Nah. I'm enjoying myself. I want to be there for the bust."
"Or the kill," said Kennett.
"Whatever . . ."
Lily pushed herself out of the chair, stretched, and tousled Kennett's hair. "Let's go look at the river," she said.
"Jesus Christ, the woman's indefatigable, and me with this heart," Kennett complained.
Lucas, vaguely embarrassed, stood and drifted toward the door. "See you guys tomorrow . . . ."
A message from Fell was waiting at the hotel: "Call when you get in, until one o'clock." He held the slip in his hand as he rode the elevator to his floor, dropped it on the bedstand, went into the bathroom, doused his face with hot water, and looked up in the mirror, the water trickling down his face.
He'd had a long relationship with a woman, the mother of his daughter, that now, when he looked back, seemed to have been based on a shared cynicism. Jennifer was a reporter, with too much time on the street, edging toward burnout. A baby, for her, had been a run at salvation.
He'd had a shorter, intense relationship with Lily, who had been struggling with the end of her own marriage; that might have been something, if they'd been in the same town, from the same emotional places. But they hadn't been, and some of the guilt of their affair still stuck to their relationship.
He'd had any number of other relationships, long and short, happy and unhappy. Most of the women he'd gone with still liked him well enough, in a wary, once-burned way; but he tended to think of them as _others,_ not Jennifer, not Lily.
Fell was one of the others. A wistful, lovely, finally lonely woman. In a permanent relationship, they would drive each other crazy. He wiped his face with one of the rough hotel towels and wandered back to the bed. He sat down, picked up the phone, looked at the receiver for a moment, then smiled. He'd felt for a year as though he were under water: quiet, placid, out of it. The New York cops were bringing him up, and Fell was fixing him in other ways. He tapped out her number. She picked it up on the second ring.
"This is Lucas," he said.
"Kennett knew it was you, but I got good mileage out of the cross-dressing thing," Fell said, without preamble. "My name was on the TV news, and it's in the _Times_ and the _Post._ That never hurts."
"I saw it . . . ."
"I'd like to find a way to thank you. Oral sex comes to mind, if I get my share," Fell said.
"Women are so _forward_ these days," Lucas said. "How quick can you get here?"
Fell brought a change of clothes with her, and they spent the evening laughing and making love. The next morning, when they were dressed, Lucas asked, "How would we find Jackie Smith?"
"Call his office," she said.
"That easy?"
"He's a hustler," Fell said. "Getting found is part of his business."
"So call him."
Smith called back in five minutes. "Aren't you guys ever going away? Can't you find out anything on your own?" he complained. "I've done everything you wanted . . . ."
"All we want to do is talk," Lucas said.
"I gave you what you wanted," Smith said again. He was angry.
"Jackie . . . ten minutes, please? Have breakfast with us or something. We'll buy."
Smith would meet them at a café outside the St. Moritz hotel, he said. They caught a cab, struggling north through the midmorning traffic, the driver with his arm out the window, whistling. The day would be hot again; already the sky was showing a whitish haze, and when they got out of the cab across from Central Park, Lucas could see the leaves on the park trees were curling against the heat.
Smith was sitting at a metal table, eating a cream cheese croissant and drinking coffee. He didn't get up when they arrived.
"Now what?" he asked, a sullen look on his face.
"We wanted to thank you—those names you gave us started a chain reaction. We've maybe got the asshole pinned down."
"No shit?" Smith looked surprised. "When'll you get him?"
"Some of the guys are betting a couple-three days. Nobody gives him more than a week," Lucas said. "But we do have something we need from you. All the small-time fences who buy from the junkies—they need to tell the dopers that Bekker'll be out looking for angel dust, ecstasy, speed. Maybe acid. And he'll kill. The guy we got to, with your help, was boosting stuff out of Bellevue, but he was also dealing dope. Bekker killed him. Cold blood. Walked up and _bam._ Killed him."
"I saw that on TV. I wondered . . ."
"That was him," said Lucas.
Smith nodded. "Okay. No skin off my butt. I'll tell everybody I know and ask them to pass the word."
"He's probably around the Village, but could be anywhere between the civic center and Central Park. That's about all we know. That's where the word's got to be," Lucas said.
"That's my territory," Smith said. "Is that all?"
Lucas glanced at Fell, then said, "No. I gotta ask you something else. You might not want to talk about it with another witness here." He tipped his head at Fell. "But if you don't mind if she stayed . . ."
Fell frowned at him, and Smith said, "What's the deal?"
"Back when I first got here, I banged up your place. Tried to get your attention . . ."
"Well, that worked," Smith said ruefully.
"Yeah. A couple of days later, I got the snot beat out of me when I was coming out of a friend's place. I need to know if that was you. Off the record. If it was, it's no problem, I swear it."
Smith dropped his croissant on the plate and laughed. "Jesus Christ, it wasn't me. I read about it, though—but it wasn't me."
"Yeah?"
"Yeah. And if you don't mind me saying so, you're the kind of guy that shit happens to, getting beat up," Smith said.
Lucas looked at Fell. "Could you hike down to the end of the block for a minute?"
"I don't know," she said, studying him.
"C'mon," Lucas said.
"Are you Internal Affairs?"
"Fuck no, I told you," Lucas said impatiently. "C'mon, take a hike."
Fell pushed back her chair, picked up her purse and stalked away.
"She's pissed," Smith said, looking from Fell to Lucas and back to Fell. "Are you screwing her?"
Lucas ignored the question: "There's a big-dog shoot-out going on. Inside the department. And I'm tangled up in it. Now. The people who jumped me might be one set of those big dogs. That's why I really need to know."
"Listen . . ."
"Just a minute," Lucas said, putting up a hand. "I want to put it to you as simple as I can. If you tell me no, it wasn't you, and I find out that it was, I'll come back and hurt you. All right? I really will, because I've gotta know the truth of this. Not knowing the truth could get me killed. On the other hand, if you say yes, it was you, there's no problem. I'll take the lumps."
Smith shook his head in disbelief, a half-smile fixed on his face. "The answer is still no. I didn't do it. I wasn't even particularly happy to see the story in the paper, because I thought you might come back on me."
Lucas nodded, and Smith spread his hands, lifted his shoulders: "I'm a businessman. I don't want any shit. I don't want any muscle around. I hate people with guns. Everybody's got a fuckin' gun." He stared off across Sixth Avenue, the traffic waiting for the light at Central Park South, then looked back at Lucas. "No. Wasn't me."
"All right," Lucas said. "So get the word out to the junkies on Bekker. You might also point out that there's a twenty-five-thousand-dollar crime-stoppers award for his capture."
Lucas turned away from Smith and walked down the street to Fell. "I wish I could read lips," she said. "I'd give a lot to know what you just told him."
"I told him why I wanted to know if those were his guys who came after me," Lucas said.
"Tell me," she said.
"No. And I'm not Internal Affairs."
They spent the day walking through the Village and SoHo, drifting in and out of shops, talking to Fell's contacts on the street, chatting with uniform cops in Washington Square, watching the street action on Broadway. They found the bookstore where Bekker had been spotted, a long, narrow shop with a narrow front window and a weathered, paint-peeled door three steps up. A sign in the door said "Open All Night, 365 Nights a Year."
The clerk who had talked to Bekker wasn't working, but happened by on his bike a few seconds after they asked for him. A thin man with a goatee and a book of poetry, he looked like a latter-day Beat, his face animated as he told them about the encounter.
"He's a good-looking woman, I'll tell you that," the clerk said. "But you can look at somebody and know what kind of book they're going to buy, and I never picked her—him—out for the one he found. Torture and shit. I thought maybe he was, like, an NYU professor or something, and that's why he bought it . . . ."
Down the sidewalk, Fell said, "I think he's real."
"So do I," said Lucas. "He saw him." He looked up at the red-brick buildings around him, with their iron stoops and window boxes full of petunias. "And he's somewhere close, Bekker is. He didn't drive any distance to get to a small bookstore. I can smell the sonofabitch."
He took her to the restaurant where Petty had been killed, sat and had Cokes, and almost told her about it.
"Not too bad a place," he said, looking around.
"It's all right," she said.
"You ever been here? Your regular precinct is around here, right?"
"Ten blocks," Fell said, poking a straw in her Coke. "Too far. Besides, this is sort of a sit-down place, not the kind of place you come to for lunch if you're a cop."
"Yeah, I know what you mean."
Late in the afternoon, while Fell browsed a magazine rack, Lucas stopped at a pay phone, dropped a quarter, and got Lily in O'Dell's car.
"Where are you?"
"Morningside Heights."
"Where's that?"
"Up by Columbia."
"I need to see you. Tonight. By yourself. Won't take too long."
"All right. How about nine, at my place?"
"Good."
When he hung up, Fell looked up from a copy of _Country Home_ and said, "So. Are you up for dinner?"
"I'm talking to Lily tonight," he said. "I'd like to come around later, though."
"I hate to see you hanging around with that woman," Fell said, dropping the magazine back on the rack.
"This is purely business," Lucas said. "And look, could you stop by Midtown and pick up those file summaries? We've been floating around all day, listening to bullshit . . . maybe something'll come out of the files."
"All right. I'll haul them over to my place . . . ."
Lily was sitting in a living room chair, her high heels in the middle of the carpet, her bare feet up on a hassock. The hassock was covered with a brocaded throw that seemed to Lucas to be vaguely Russian, or Old World. She was sipping a Diet Coke, tired smudges under her eyes.
"Sit down. You sounded tense," she said. "What happened?" Her head was back, her dark hair a perfect frame around her pale oval face.
"Nothing happened, not today, anyway. I just need to talk to you," he said. He perched on the edge of her other overstuffed chair. "I need to know about you and Walter Petty—your relationship."
She leaned farther back in the chair, wiggled once to settle in, laid her head back, and closed her eyes. "Can I ask why you need to know?"
"Not yet."
She opened her eyes and looked at him carefully and said, "Robin Hood?"
"I'm not sure. What about Petty?"
"Walt and I went back as far as you can go," Lily said, her eyes unfocusing. "We were born on the same block in Brooklyn, sort of middle-class brownstones. I was exactly one month older, to the day. June first and July first. His mother and mine were friends, so I suppose I first laid eyes on him when I was five or six weeks old. We grew up together. Went to kindergarten together. We were both in the smart group. Someplace along the way, sixth or seventh grade, he got interested in math and science and ham radio in that geeky way boys do, and I got interested in social things. After that we didn't talk so much."
"Still friends, though . . ."
She nodded. "Sure. I'd talk to him when I saw him around the block, but not at school. He was in love with me for most of his life. And I guess I loved him, you know, but not sexually. Like a handicapped brother, or something."
"Handicapped?"
She carefully set the glass on the table and said, "Yeah, he was socially handicapped. Walked around with a slide rule on his belt, his table manners went from bad to worse, he got weird around girls. You know the type. Sort of ineffectual, nonphysical. Really nice, though. Eager . . . too eager."
"Yeah. A dork. A nerd. The kind of kid that gets shredded by girls."
"Yes. Exactly. The kind that gets shredded," she said. "But we were friends . . . . And whenever I needed something done—you know, get an apartment painted, or help fixing something—I could call him up and he'd drop everything and be there. I took him for granted. He was always there, and I assumed he always would be."
"Why'd he become a cop?"
" 'Cause he could. It was a job you could get with a test and with family connections. He was brilliant on tests and had the connections."
"Was he a good cop?"
"He was terrible in uniform," she said. "He didn't have that . . . that . . . cold spot. Or hot spot. Or whatever it is. He couldn't get on top of people—you ought to know about that."
"Yeah." Lucas grinned. "I don't know if it's hot or cold, though. Anyway, Petty . . ."
"So he was terrible on the street and they moved him inside. He was working guard details and so on. Then they tried him on dope. And Jesus, he was something else. I mean nobody, _nobody_ would believe he was a cop. He'd make a buy and the backup would drop on the dealer, and they still wouldn't believe it. This _dork_ couldn't be an undercover cop. Sometimes even the judges didn't believe it. Anyway, that's about the first job he ever did really well at; he was a bit of an actor. Then he got interested in investigation, in crime-scene processing. He was good at that, too. The best. He'd go into a crime scene and he'd see _everything._ And he could put it together, too. Then computers came along, and he was great with computers." She laughed, remembering. "Suddenly, the guy who fucked up everything, the nerd as big as the moon, was a hot item. And he was still good old Walt. When you needed your apartment painted, there he was. He had this great open smile, completely . . . geeky, but honest. When he looked happy to see you, he was happy to see you; he'd just light up. And if he got angry, he'd go off and start yelling, and then he'd maybe start crying or something; or you thought he would . . . ."
Lily's lip was trembling, and she dropped her feet off the hassock and dropped her head.
"How'd he get the job looking for Robin Hood?"
"He knew computers and he'd worked with O'Dell, and we swung it for him. He could help us, and it was a chance for him to break out. And maybe I had something to do with it—he'd be working with me. Like I said . . ."
"Yeah. I know exactly what you mean."
"Sounds like arrogance, or vanity."
Lucas shook his head. "Not really. Just life . . . You think he got close to Robin Hood?"
"He must have. Jesus, when he was killed, I couldn't stop crying for a week. I really . . . I don't know. There was no sexual impulse at all, but when I thought of him over all those years, that puppy-dog quality, that he loved me . . . It was like . . . I don't know. I loved him. That's what it came to."
"Huh." He was watching her, his elbow on the arm of the chair, one finger at his chin.
"So what's this all about?" she asked. The weariness had slipped from her voice, and she looked up, intent.
"You and O'Dell are running me as some kind of lure," Lucas said. "You're dragging me out in front of whoever your targets are. I need to know who you think they are."
After a long moment of silence, she said, "Fell. As far as I know, that's it."
"Bullshit."
"It's not bullshit," she said. "She's all we've got."
"That can't be right."
"It is."
"You know everything that O'Dell is doing?"
"Well, yes, I mean I schedule for him . . . I suppose he could run something on the side . . . ."
There was another moment of silence, then Lucas said, "I'm afraid you're betraying me."
She was offended, angry. "God damn it."
"I know you are—or somebody is. O'Dell for sure, and you're with O'Dell . . . ."
"Tell me about it," she said, sitting back again.
Lucas looked her over and said, "First of all, Fell's not involved."
"Why not?"
"I just know, and I'm not wrong," Lucas said.
"Lucas, instincts or no instincts, the goddamn records aren't lying about this," Lily said. "She's all over the place."
"I know. She's an alarm."
"What?"
"She's a trip wire," Lucas said. "Working the jobs she has, in Burglary, and as a decoy, she knows half the assholes in Midtown. So Robin Hood used her as a reference and picked on assholes that she knew. Then they watched her. If anybody got close, they'd get close to her first . . . ."
"I don't know." Lily was shaking her head. She didn't believe it.
"It'd have to be a tough sonofabitch to set that up," Lucas continued. "As soon as you pulled her off her regular job and put me next to her, the alarm went off. Petty's been killed, the official investigation seems to be dead in the water—and here comes Lily Rothenburg and the department's Svengali, towing me along behind. And you stick me next to Fell. They never bought the Bekker thing: they've been reading us like a book."
"Who?"
Lucas hesitated. "I'm tempted to say Kennett."
"Bullshit." Lily shook her head. "I'd know. In fact, I asked him. He doesn't even think there is such a group."
"But we know there is. And I'm still tempted to say Kennett. O'Dell put me right up against Fell and he put me right up against Kennett. It's possible that O'Dell _knows_ it's Kennett, but doesn't have the proof."
Lily thought it over, staring at him. "That's . . ."
"Bizarre. I agree. And of course, there're other possibilities, too."
"That it's me?" She smiled a small and frosty smile.
"Yeah." Lucas nodded. "That's one of them."
"And what do you think?"
He shook his head. "It's not you, so . . ."
"How do you know it's not me?" she asked.
"Same way I know it's not Fell—I've seen you operate."
"Thanks for that," Lily said.
"Yeah . . . which brings us to the last possibility."
"O'Dell?"
"O'Dell. He has access to everything he needs to organize the group. He knows everybody on the force, and he probably could pick out likely candidates for his hit teams. He has the computer files to pick out the assholes, and to set up Fell as an alarm . . . ."
"There's a hole," Lily said quickly. "He's so high up he wouldn't need an alarm . . . ."
"Internal Affairs—he might not know about Internal Affairs investigations."
She bit her lip. "Okay. Go ahead."
"Since Petty was a computer maven too, maybe computers led him to O'Dell. Whatever it was, for whatever reason Petty got hit, O'Dell was right there to manage the investigation. Kept it out of Internal Affairs . . ."
"Said it was too political," Lily said thoughtfully.
"Yeah. Then he pulls me into it, produces Fell, and he puts me up against Kennett. And you know what? Fell and Kennett are all I've got—all that paper you gave me, the regular investigation, the reports. It's all bullshit. It's all a stone wall. It looks impressive, but there's nothing in it."
"Why would O'Dell pick on Kennett?"
"Because Kennett's going to die," Lucas said bluntly. "Suppose he gets everything pointed at Kennett, and then Kennett . . . dies. Natural causes, a heart attack. If there was an agreement that Kennett was it, the investigation would die and the real organizer would be clear."
Lily, pale as notebook paper: "He couldn't have . . . I don't think."
"Why not?"
"I don't think . . . I don't think he's brave enough. Physically. He'd be thinking about prison."
"That all depends on how he's set it up. Maybe his shooters don't know him."
"Yeah, but remember—if O'Dell is it, he wouldn't have to give you Fell. If Fell's an alarm, I mean, he'd know what you were here for."
"Yeah. And he'd know that Fell would get me exactly where she has: nowhere. And at the same time, lend a touch of truth to the whole business. Fell did know all those dead guys. Besides, with Petty talking to both of you, and Fell popping out of the computer, there was no way to get her back inside . . . ."
"Maybe," she said.
"How'd you meet Kennett?" Lucas asked abruptly.
"In the intraconference meetings."
"As O'Dell's assistant?"
"Yes."
"Did O'Dell feed you to him?" Lucas asked.
"Jesus, Lucas," she said.
"Did he? I mean, he knows both of you. Could he have figured . . ."
"I don't know. They don't like each other, you know." Lily stood and turned in place, like a dog trying to make a bed more comfortable. "You know, you've put this whole tissue together without a single goddamned fact . . . ."
"I've got one interesting, surprising, generally unknown fact," he said; and it was his turn to produce a wintry smile.
"What?"
"I know that O'Dell's trying to frame Kennett. I know _that_ for sure. The question is, is he doing it because Kennett's guilty and it's the only way to get him? Or because he's looking for a scapegoat?"
"Bullshit," she said, but he could see the shock in her eyes.
"I found Red Reed in Charleston, South Carolina," he said. "He's a friend of O'Dell's, from Columbia . . . ."
And then he told her most of the rest of it, except for the curious thing Mrs. Logan had said, when they interviewed her in the apartment below Petty's.
## CHAPTER
24
Lily listened as Lucas called Fell, watched his face, watched him smiling, turning away, setting up a date. Lucas left, hurrying, and she stood at the window with her purse, watching him. He flagged a cab, and just before he got in, looked up and saw, pointed at her purse, waved.
Then he was gone.
She walked through the apartment, touching things, with the sense of something ending, with a sense of dread.
Kennett? No. But O'Dell was unthinkable too. Could O'Dell have coldly executed his own man . . .
Finally, she picked up the phone and punched in the number for Kennett's boat. He picked it up and said, "Lily."
Pleased, she said, "How'd you know it was me?"
"I think it might be love," he said. "Are you feeling lonely?"
"You're reading my mind."
"The river's beautiful tonight . . . ."
The river was quiet, smelling of mud and oil and salt. Halyard hardware tinkled against the aluminum masts. A late-night squall was rolling off the coast far to the northeast, and they could see the lightning in the sky far beyond the lights of Manhattan.
As Lily and Kennett made love, she had a moment of absolute clarity, could hear the Crash Test Dummies' song "Superman" roll mournfully out of a nearby boat, muted by the ten thousand unidentifiable cheeps and knocks of the marina.
Later, in the cockpit . . .
"Jesus, I'm sitting here bullshitting and you're sitting there crying," Kennett said quietly. He reached across and thumbed a tear off her cheek. "What's all this about?"
"I was just looking across the river, thinking how pretty it was, how good it feels. Then I thought about Walt, about how he'd never see it again."
"Petty?"
"Yeah. God damn it."
"The guy has a strange hold on you, m'dear," Kennett said, trying to keep his voice light: an invitation to talk.
"You know why?" she asked, taking up the invitation.
"Why?"
"Because we were so goddamn mean to him, that's why. Us girls, in school. Lucas got me thinking about it . . . ."
"It's hard to see you as mean," Kennett said.
"I didn't think about it at the time. The thing about Walt was, he'd do anything for you. He was always so _eager._ And when we were in school—and even after that, on the force—we paid him back by laughing about the way he dressed, and the way he acted, and all those pens he used to carry around. We made him be a clown and he wasn't a clown; but whenever he tried to be serious, we wouldn't let him. We hurt him. That's what I was thinking about, the times I know we hurt him—girls, in high school—that hurt look on his face when he'd try to do something, try an approach and we'd laugh in his face. He never really understood . . . . Oh, God."
Suddenly, she was sobbing and Kennett patted her on the back, helplessly. "Jesus, Lily . . ."
A moment later she said, her voice clearing, "You're a Catholic. Do you believe in visions? You know, like the Virgin Mary and all of that, talking to shepherds?"
"I'd want to see it myself," Kennett said wryly.
"The thing is, I keep seeing Petty . . . ." She laughed, a short, sad laugh, and poked him. "No, no, no, I don't see him floating around my room, I see him in my mind . . . ."
"Whew."
"But the thing is, it's so clear. Walt running down the street, and his hair plastered down and his ears sticking out . . . Jesus Christ. Walt was the only guy who ever loved me and didn't want anything from me. No sex, no kids, no favors, just me being there and he was happy."
Kennett found nothing to say, and they sat there, their feet up, watching the dark river. After a while, Lily began to cry again.
## CHAPTER
25
Lucas called Fell from Lily's, apologizing for the late hour.
"I was going down to the tavern," she said. "Why don't you meet me . . . ."
He flagged a cab, Lily watching from her window, smiling down at him. He waved, and she lifted her purse in her left hand, slipped her right inside the gun tote. _Remember the last time?_
At the tavern, Lucas pulled a twenty out of his Muskies Inc. money clip and tipped the driver two dollars for the eight-dollar ride. Fell was in the back booth, a beer on the table with a bowl of peanuts. She was reading a free newspaper.
"Hey," he said, slipping into the booth.
"Hi. Any developments at Rothenburg's?"
"No . . ."
"Good," she said.
Lucas shook his head. "Jesus." And then: "I gotta get a beer." He waved at a waitress, pointed at Fell's glass and gave her a victory/two sign. While they waited, a swarthy man in a light-blue sport coat and khaki slacks, a glass of dark beer in his hand, wandered up to the table and said to Fell in a bad imitation Bogart, "Howdy, shweet-heart. Sheen your name in the public prints."
"Hey, Tommy. Sit down." Fell patted the seat beside her, then pointed her trigger finger at Lucas. "That's Lucas Davenport, who's a cop."
"I know who he is," Kantor said, dropping into the booth. "But somehow I got left off the invitation list for the Welcome to New York interviews."
"And Lucas," Fell continued, "this is Tommy Kantor, who's a columnist for the _Village Voice._ . . ."
They talked about the case for a while, and Kantor attracted the attention of a free-lance magazine writer and his girlfriend. They pulled up a chair and ordered a pitcher of beer. Then a TV producer stopped by and began talking to Fell.
"You'd make a good piece," she told Fell.
"I'd certainly agree with that," Lucas said, straight-faced.
"Fuckin' Davenport . . ." Fell said.
They got back to Fell's apartment at two o'clock, spent ten soapy minutes in the shower, dropped into her bed.
"That was fun, talking to those people," Lucas said. "As long as your friend Kantor doesn't get us in trouble."
"He takes care of sources," Fell said. "It'll be okay. I'm surprised you get along so well with media people . . . ."
"I like them, mostly," he said. "Some are a little stupid and half of them would kill for two dollars, but the good ones I like."
"You like _this_?" she asked.
"Ooo, I think I do," he said. Then: "I'm sure of it."
He came out of the shower the next morning, rubbing his hair dry with a terry-cloth towel, and heard Fell's voice from the living room. She came down the hall to the bedroom as he was pulling on his underwear. She was still naked and stood on her tiptoes to kiss him.
"I just talked to Carter. Not a thing, nada."
"All right. Did you bring those files?"
"In the front room, on the floor," she said.
"I'd like to sit around and read for a while, then maybe go back and change clothes. I don't know, I'd like to be there when they get him . . . ."
"Bullshit. You'd give your left nut to get him yourself. So would I."
"You'd give my left nut?" he asked, appalled.
"Well . . . you want a bagel with chive cream cheese and some juice?"
"Yeah, as a matter of fact."
They read the files and talked, and sometime after one o'clock Lucas chased her back into the bedroom, and they didn't make it back out until two.
"I'm going back to the hotel to change," he said, pulling on his jacket. "Why don't we get together at Midtown. Like four-thirty, for the daily roundup."
"All right . . ."
He looked at the floor by his feet, at a Xerox copy of the crime-scene photograph of Whitechurch, dead in the hospital. The few pitiful twenties stuck out from under his body like a comment on greed.
"Change oxen in midstream and you'll come to a bad end," he said.
"What?"
"An old English proverb my mom used to tell me," Lucas said.
"Bullshit," she said.
"You're calling my mom a liar?"
"Get out of here, Davenport. See you at four-thirty."
He took the elevator to the lobby, nodded at a guard who knew a one-night stand when he saw one, spotted a cab pulling up to the curb to drop a passenger, stopped and slapped his coat pocket where his wallet was.
"Dammit," he said.
"Hub?" The guard looked up from his desk.
"Sorry. Not you . . . I forgot something upstairs."
He went back up, knocked on the door. Fell, wrapped in a robe, let him in. "You got twenty bucks you can loan me?" he asked. "I got like two dollars left after last night. All the traveler's checks are at the hotel."
"Oh, jeez . . ." She went to her purse, opened it, took out a billfold. "I've got six bucks," she announced. Then she brightened and dug further. "And a cash card. There's a machine down the block. I'll trust you with my code and change it when you skip on me."
He looked at the cash card, looked down past it to the floor, at the Xerox of Whitechurch, the twenties under his body. The money, the money. Bekker.
"Get dressed," Lucas snapped. "Hurry the fuck up."
Three twenty-dollar bills had been found around and under Whitechurch's body. They drew the money from the evidence locker, under the watchful eye of the custodian.
"Consecutive?" Fell whispered. She was excited, barely controlled.
Lucas scanned the numbers, rearranged the bills on the countertop. "Two of them," he said. He took the numbers down on a notepad. "Let's go talk to the feds."
Terrell Scopes of the Federal Reserve had a procedure for everything, including the dispensing of information about serial numbers. "I can't just have people come in here . . ." He waved, a wave that seemed to suggest that they didn't quite meet a standard. Lucas was rumpled. Fell's hair was beginning to go haywire, standing around her head in a halo.
"If we take several hours to get the data and Bekker cuts the heart out of somebody, your picture'll be on the front page of the _New York Times_ right along with his," Fell snarled, leaning across his desk.
Scopes, naturally pale, went a shade paler. "Just a minute," he said. "I'll have to make some inquiries."
After a while he came back and said, "Citibank . . ."
Citibank was more cooperative, but the process was a long one. "The money came out of a machine on Prince, all right, but exactly when, or where it went, that'll take a while to figure out," said a round-faced banker named Alice Buonocare.
"We need it in a hurry," said Lucas.
"We're running it as fast as we can," Buonocare said cheerfully. "There's a lot of subtraction to do—we have to go back to a known number and then start working through the returns, and there's a lot of stuff we have to do by hand. We're not set up for this kind of sorting . . . and there are something like twenty thousand items . . . ."
"How about the pictures?"
"They're not really very good," Buonocare confessed. "If all you know is that he's got blond hair, there are probably a thousand blondes on the tape . . . . It'd be easier to do the numbers, then confirm with the pictures."
"All right," Lucas said. "How long?"
"I don't know: an hour, or maybe two. Of course, that's almost quitting time."
"Hey . . ." Lucas, ready to get angry.
"Just kidding," Buonocare said, winking at Fell.
Three hours. A mistake was found halfway through the first run, a question of which numbers went where, and another machine on Houston Street.
"All right," one of the computer operators said at six o'clock. "Give us another twenty minutes and we'll have it down to one person. If you want to look right now, I can give you a group of eight or ten and it's ninety percent that he's in that group."
"How about the photos?"
"We'll get the tape up now."
"Let's see the ten accounts," Buonocare said.
The programmer's fingers danced across the keyboard and an account came up on the green screen. Then another, and another and more. Ten altogether, six men, four women. Two accounts, one man, one woman, showed non-Manhattan addresses, and they eliminated them.
"Can we get account activity on the other eight? For the last two months?" Buonocare asked over the shoulder of the computer operator.
"No problemo," he said. He rattled through some keys, and the first account came up.
"Looks routine . . ." Buonocare said after a minute. "Get the next one."
"Better find it in a hurry," Fell said. "I'm about to pee my pants."
Edith Lacey's account was the fifth one they looked at. "Oh-oh," Buonocare said. To the computer operator: "Get the rest of this up, go back as far as you can."
"No problemo . . ."
When the full account came up, Buonocare reached past the computer operator and pressed a series of keys, then paged down through an extensive account listing. After a moment, she ran it back to the top and turned to Lucas and Fell.
"Look at this: she started with a balance of $100,000 six weeks ago, and then started pulling out the max on her bankcard, five hundred a day, just about every day for a while. Even now, it's three or four times a week."
"That could be him," Lucas said, nodding, excited. "Let's get a picture up. You've got a name and address?"
"Edith Lacey . . ."
"In SoHo. That's good, that's right," Fell said, tapping the screen.
"How about the video . . . ?"
"Let's get the reference numbers on those withdrawals . . ." Buonocare said. She wrote the number on a scratch pad and they carried it to the storage. The right cassette was already in the machine, and Buonocare ran it through, looking at the numbers . . . .
"Here," she said.
The screen showed a blonde, her face down.
"Can't tell," Fell said. "I swear to God, I'm gonna pee in my pants."
"Let's try another withdrawal in that sequence," Buonocare said.
She ran the tape, stopped, started, searched. Found another blonde.
"Motherfucker," Lucas said, looking at the screen. "Nice to see you again, Mike."
"That's him?" Fell asked, peering at the screen. "He's so pretty."
"That's him," Lucas said.
Bekker was smiling at the lens, his blond hair pulled demurely away from his forehead.
## CHAPTER
26
Bekker awoke at noon. He wandered about the apartment, went to the bathroom, and stared at himself. Pretty. Pretty blonde. Too late for pretty blonde.
He cried, sitting on the edge of the tub, but he had to do it. He shaved his head. Hacked his fine silken hair to stubble with a pair of orange-handled scissors from Mrs. Lacey's sewing box, lathered it with shampoo, scraped off the stubble with a safety razor. Cut himself twice, the blood pink in the lather . . .
Sigh.
He found himself in front of the mirror, dried soap around his ears, hair. Gone. The tears came again, in a rush. His head was far too small, and sickly white, like a marble. Where was Beauty?
He examined himself with the eye of an overseer, the Simon Legree of inspections. Bald. Pale. No good. Even in the Village, the scalp pallor would attract the eye, and the facial makeup would be obvious.
The scars—the scars would give him away. He touched his face, felt the furrowed, marbled flesh. A new role, that's what he needed. He'd thought to cut his hair, shift back to a male role, but that wouldn't work. Besides, women were allowed a greater latitude of disguise. He'd go back to the wigs he'd worn before his own hair grew out.
Bekker strode through the apartment, headed for the stairs, stopped to touch the cloud of spiders that hung over his desk in the outer apartment. So fine, so pretty . . .
Go. Get the wig, get dressed—he hadn't bothered to dress. Clothes seemed inconvenient and restrictive. He marched now, directed by the PCP, upright and dignified, then he was suddenly aware of his penis, bobbling along like an inconveniently large and flaccid nose, doing a color commentary on his dignity. Bekker pressed his penis to his thigh, but the rhythm of the march was broken . . . .
A new gumball dropped. From when? The fifties? A comedian on _The Ed Sullivan Show?_ Yes. A small man looking into a cigar box, talking to a voice inside . . . _Okay? Okay._ Was that the line? Yes.
Bekker, passing the kitchen, swerved, went in. Opened the refrigerator and peeked inside: Have a Coke, Mr. Bekker. Thank you. I will. _Okay? Okay._ He slammed the refrigerator door like the comedian and howled with laughter. _Okay? Okay._
Really funny . . . He howled . . . .
Coke in hand, he staggered back to the television, turned to CNN, and watched for a few minutes. He'd been on one of the news shows in the morning, with the pictures of the Carson woman; they'd ridiculed him, said the halos from Carson had been finger-press points on the photo paper. What did that mean? Was that methodology? He had a hard time remembering anymore . . . .
He watched, hoping to see the report again, but they'd cut him out of the news cycle.
He went downstairs, naked and barefoot, stepped carefully through the shambles of the first-floor shop, and down into the basement. Found the dark wig, with the pixie cut. Carried it back up, to the bathroom, put it on. It was warm on his head, like a fur piece, and scratchy. But it looked good. He'd have to do something about his eyebrows, shade them, and his lashes. Maybe something to tone his face . . .
Mrs. Lacey had been too old for sophisticated makeup, had been satisfied with a pinkish rouge to make two little pink spots on her cheeks, like Ronald Reagan's. But she had an eyebrow pencil. He found the pencil, came back to the mirror, wet it with his tongue and began feathering it through the lashes. A new face began to form in the mirror . . . .
He ventured out at five-thirty, tentative, wary, the day still bright, and turned toward Washington Square. He was unused to the sunlight, and squinted against it, his speed-hyped vision dazzled by the color and intensity. He carried his handbag and an old newsprint drawing pad he had found in one of Mrs. Lacey's cupboards.
Not much foot traffic, not north and south. He stayed on the shadier side of the narrower streets, head down. Dark hair, dark eyebrows, dark blouse, jeans, gym shoes. A little dykey. A little too tough for a woman. An attitude.
During his early reconnaissance of the city, he'd seen some action around the square. Dealers drifting through. Baggies and cash. He felt the plastic box in his jeans pocket, the tabs rattling inside. Six left, six between himself and . . . He couldn't think about it. He had five thousand in cash in his purse, and the pistol, just in case.
He needed some luck.
• • •
Oliveo Diaz had ten hits of ex and another ten of speed, and maybe a couple of hours to sell it. Party that night; he could use the cash to pick up some coke for himself. Coke was a mellower high than the speed. With enough speed, Oliveo felt that he could go anywhere. With cocaine, he'd already arrived.
Oliveo crossed the south side of the square, saw Bekker sitting on a concrete retaining wall, sketching. Looked nice, from a distance, with the inky black hair, like maybe a PR. Closer, and he thought, maybe Irish, black Irish with the pale skin.
Bekker paid no attention to him, his face down in the sketch pad, a pencil busy in his hand. But watching . . .
"Hey, Oliveo, doood . . ."
Oliveo turned, flashed the automatic smile. Some guy named Shell. Young white guy with a battered forehead, hazy blue eyes and a Mets hat with the bill turned backward. Oliveo had a theory that a guy's intelligence could be determined by how far around his head the bill was turned. Backward was a complete fool, unless he was a baseball catcher. Shell's hat was backward, and he said again, "Hey, doood," and he lifted a hand for a cool five.
"Shell, my man, what's happenin' . . . ?" Oliveo said. Shell worked in a tire-recap place, had cash sometimes.
"You servin'?" A quick look left and right.
"Man, what you need?" The smile clickin' on again. Oliveo thought of himself as a pro, a street Mick Jagger, smile every ten seconds, part of the act.
"Gotta get up, man . . ."
"I got ten hits of really smooth shit straight from Miami, man . . . ."
• • •
Bekker sat on the wall and drew the fire hydrant; drew it well, he thought. He'd learned drawing techniques in medical school, found them useful as a pathologist. They made structure clear, simple. He struggled to keep the drawing going as he watched Oliveo chatting with the white kid, watched them circle each other, checking for cops, and finally a flash of plastic.
Bekker looked around. There _were_ cops in the square, but on the other side, near the arch. Three blue Plymouths parked side by side, the cops sitting on the hoods or leaning on the fenders, talking. Bekker picked up his purse and, as the white guy peeled away from Oliveo, sauntered over.
"Servin'?" he squeaked.
Oliveo jumped. The woman with the art pad, her head down. He couldn't see her face very well, but he knew he'd never dealt to her. She was wrong, something wrong. A cop?
"Get the fuck off me, man," he said.
"I've got a lot of cash," Bekker said, still squeaking. He sounded like a mouse in his own ears. "And I'm desperate. I'm not a cop . . . ."
The word "cash" stopped Oliveo. He _knew_ he should walk away. He knew it, had told himself, don't sell to no strangers. But he said, "How much?"
"A lot. I'm looking for speed or angels or both . . . ."
"Fuckin' cop . . ."
"Not a cop . . ." Bekker glanced up the street, over at the cop cars, then put his hand in the bag and lifted out an envelope full of cash. "I can pay. Right here."
Oliveo looked around, licked his lips, then said, "What you look like, mama?" He reached out, grabbed Bekker under the chin and tried to lift his face. Bekker grabbed his arm at the wrist and twisted. There was muscle there, testosterone muscle. As he pushed Oliveo away, his head came up, his teeth bared, eyes wide.
"Motherfucker . . ." Oliveo said, backing away, sputtering. "You're that dude."
Bekker turned away, started across the street, half running, mind twisting, searching for help, for an answer, for anything.
Behind him, Oliveo had turned toward the cop cars across the square. "Hey," he screamed. He looked from the cops to Bekker, then at the cops again, then dashed toward them, yelling, waving his arms. "Hey, hey, that's him, that's him . . . ."
Bekker ran. He could run in the gym shoes, but there were a lot of cops, and if they came quickly enough, and if they asked about a woman running . . .
A bum stood at the mouth of an alley, picking through a garbage can. He wore a crumbled hat and a stained army coat, ankle length.
A half-brick sat on the sidewalk, a remnant of concrete lapped over it like frosting on a piece of carrot cake.
It was a narrow street, the closest people a block away, not looking.
Bekker snatched the brick off the street, still running. The bum looked up, straightened, leaned away, astonished when Bekker hit him squarely in the chest. The bum pitched over the garbage can and went down into the alley, on his back. "Hey," he groaned.
Bekker hit him between the eyes with the brick, then hit him again. Hovered over him, growling like a pit bull, feeling his blood rising . . .
A siren, and another.
He stripped the hat and trench coat from the bum, pulled the trench coat over the purse, stripped off the wig, pulled the hat down low on his head. The bum blew a bubble of blood. Still alive. Bekker lurched back to the mouth of the alley, trying on the new persona, the mask of beggary . . . .
Behind him, a gargling sound. He half turned; the bum was looking at him, one good eye peering brightly out of a ruined face. The bum was dying. Bekker recognized the gargle. Something cold, distant and academic spoke into his mind: cerebral hemorrhage, massive parietal fracture. And that eye, looking at him. The bum would die, and then he'd be back, watching . . . . Bekker looked both ways, then hurried back to the bum. Pocketknife out, quick jabs; eyes gone. The bum moaned, but he was going anyway.
The brick was by the bum's head, and Bekker picked it up and jammed it in his pocket. Good weapon. A gun was too noisy. But he groped for the gun inside the bag and transferred it to the pocket.
Into the street. Six blocks. He saw a cop car go by, screech to a stop at the intersection, the cops looking both ways out the windows, then go on. The coat stunk: dried urine. The smell clogged his throat, and he imagined fleas crawling onto him. More sirens, cops flooding the neighborhood. Bekker hurried . . .
Turned onto Greene, tottering, a drunk, his shabby coat dragging on the pavement. A woman coming. Closer, same side. Bekker changed to the other side of the street. His vision wavered, changed tenses: Approaching Lacey building. Sirens in the distance, but fading. Woman goes to Lacey building door . . .
_What_ . . .
The panic gripped him for a moment. Confused. What did she want? Blank-faced buildings looking down. Gumball drops. Red one, loading anger. They would do this to him, a man of talent. The woman was half turned toward him, head cocked.
A distant voice, in the back of his head: Bridget. Bridget Land. Come to visit . . .
He straightened, walked back across the street, away from her, and she put a key in the front-door lock and turned it, pushing the door open. Bridget Land, he'd forgotten about her . . . . She must not know.
She pushed the door open, her shoulders rounded, aged, straining with the effort, then stepped up and inside. Bekker, caught by anger and opportunity, began moving. There was no space or time, it seemed, and he hit the door, smashed inside, and hit her.
He was fast, angel-dust fast, quicker than a linebacker, smacking her with the brick full in the face. She went down with a strange, harsh croak, like a wing-shot raven.
Bekker, indiscreet, beyond caring, slammed the door, grabbed her by the hair and dragged her to the stairs, and down.
He forgot the bum's clothes, and paid no attention to the woman, yipping like a chihuahua with a bone in its throat. He dragged her to the room, strapped her down. Her legs started to work now, twitching. He wired the silencer into her mouth, working like a dervish, hovering . . . .
## CHAPTER
27
Buonocare, the banker, ran the photo tape through two more withdrawals. Bekker posed in all three, a startling feminine beauty coming through despite the rough quality of the tape.
"Jesus, I wish I looked that good," Buonocare said. "I wonder who does his hair."
"Gotta call Kennett," Fell said, reaching across the desk to pick up a phone.
"No." Lucas looked into her eyes, shook his head. "No."
"We've gotta . . ."
"Talk to me outside," Lucas said, voice low.
"What?"
"Outside." Lucas looked at Buonocare and said, "There's a security thing here, I'm sorry I can't tell you . . . ."
Fell got her purse, Lucas his coat, and they half ran to the door. "Will I see it on the news?" Buonocare asked as she escorted them past a security guard to the front door.
"You'll probably be _on_ the news, if this is him," Fell said as the guard let them out.
"Good luck, then. And see you on TV," Buonocare said. "I wish I could come . . ."
Outside, it had begun to rain, a warm, nasty mist. Lucas waved at a taxi, but it rolled by. Another ignored him.
Fell grabbed his elbow and said urgently, "What're you doing, Lucas? We've gotta call now . . . ."
"No."
"Look: I want to be there too, but we don't have time. With this traffic . . ."
"What? Fifteen minutes? Fuck it, I want him," Lucas said.
"Lucas . . ." she wailed.
A cab pulled to the curb and Lucas hurried over, three seconds ahead of a woman who sprinted from a door farther up the street. He hopped in, leaving the door open. Fell was behind him, still in the street. "Get in."
"We gotta call . . ."
"There's more going on here than you know about," Lucas said. "I'm _not_ Internal Affairs, but there's more going on."
Fell looked at him for a long beat, then said, "I knew it," and climbed in the cab. As the cab pulled away, the woman who'd run for it, back in the doorway, gave them the finger.
They inched silently uptown through the nightmare traffic, the rain growing heavier. Fell was tight-lipped, agitated. The cab dropped them on Houston, Lucas paid. A squad car rolled by, the cops looking carefully at Lucas before going on. They dodged into a convenience store, damp from the misty summer rain.
"All right," said Fell, fists on her hips. "Let's have it."
"I don't know what's going to happen, but it could be weird," he said. "I'm trying to catch Robin Hood. That's why they brought me here, from Minneapolis."
Her mouth dropped open. "Are you nuts?"
"No. You can either come along or you can take a hike, but I don't want you fuckin' this up," Lucas said.
"Well, I'll come," she said. "But Robin Hood? Tell me."
"Some other time. I gotta make a call of my own . . . ."
Lily was with O'Dell, just coming off the Brooklyn Bridge into Manhattan, ten minutes from Police Plaza.
"Have you heard?" she asked.
"What?"
"Bekker was spotted at Washington Square, but took off. This was around three o'clock. We've got people all over the place, but nothing since . . . ."
"That sounds right, because I think we know where he is. Fell and me. And it's up in SoHo."
"What?" And he heard her say, "Lucas says he's got Bekker."
O'Dell's voice replaced Lily's. "Where are you?"
"We're at Citibank and we're stuck here. I think Bekker's holed up with an old lady in SoHo, but I'm not sure. I'm going up there to take a quick informal look around before I call in the troops. I just wanted Lily to know, in case something misfires . . . ."
"Besides, if you called now and you're stuck downtown, Kennett would get all the credit for the bust," O'Dell said with his wet chuckle. "Is there any possibility that what you've done, whatever it is, has tipped off Bekker?"
"No. But it'll take us a while to get up there; it's raining here, and cabs are impossible."
"Yeah, it's raining here, too . . . . Okay, go ahead. But take care. Just in case there's a problem, why don't you give me the address, and I'll get Lily to start a search warrant. That'll help explain the delay, why you didn't call it in."
"All right . . ." Lucas gave him the address, and Lily came on the line. "Careful," she said. "After your . . . look around . . . give us a ring. We'll have the backup waiting."
Lucas hung up, and Fell asked, "All right—what's going on?"
"We're gonna surveil for a while . . . ."
"Surveil what?" Another cop car rolled by, and again they got the look.
"This Lacey woman's building, for a start. Bekker knows me, I don't want to go right up front . . ."
"I know where we can get a hat," Fell said. "And it's on the way . . . ."
They dodged from doorway to canopy, staying out of the rain as much as they could. Fell finally led Lucas into a clothing store that apparently hadn't changed either stock or customers since '69. Every male customer other than Lucas was bearded, and three of the four women customers wore tie-dye. Lucas bought an ill-fitting leather porkpie hat. In the mirror, he looked like a hippie designer's idea of an Amazon explorer.
"Quit grumbling, you'd look cute in the right light," Fell said, hurrying him along.
"I look like an asshole," Lucas said. "In any light."
"What can I tell you?" she said. "You ain't posing for _Esquire._ "
The rain had slowed further, but the streets were wet and slick, stinking of two centuries of grime emulsified by the quick shower. They found Lacey's building, cruised it front and back. The back wall was windowless brick. A weathered shed, or lean-to, folded against the lower wall. The gate in the chain-link fence had been recently opened, and car tracks cut through the low spotty weeds to the shed.
Lucas walked to the edge of the lot, where he had the sharpest angle on the shed. "Look at this," he said.
Fell peered through the fence. The back end of a rounded chrome bumper was just visible inside the shed. "Sonofabitch, it's a Bug," she breathed. She grabbed his arm. "Lucas, we gotta call."
"Lily and O'Dell are taking care of it," he said.
"I mean Kennett. He's our supervisor. Christ, we're cutting out the boss . . . ."
"Soon," Lucas promised. "I want to sit and watch for a few more minutes."
They walked around front, and Lucas picked out a store a hundred feet up the street from Lacey's, on the opposite side, an African rug-and-artifact gallery. The owner was a deep-breasted Lebanese woman in a black turtlenecked silk dress. She nodded, nervous, and said, "Of course," when they showed their badges. She brought chairs and they sat at an angle to the window, among draperies and wicker bookcases, watching the street.
"What if he goes out the back?" asked Fell.
"He won't. There're cops all over the place. He's holed up."
"Then what are we waiting for?"
"For some guys. Robin Hood and his merry men. If nothing happens in a half-hour, we go in . . . ."
"Would you like some cookies?" the Lebanese store owner asked, a touch of anxiety in her voice. She was twisting her hands, and looked, Lucas thought, remarkably like the wicked-witch stepmother in _Snow White,_ if he had his Disney movies right. "Baklava, maybe . . . ?"
"No, thanks, really," Lucas said. "We're fine. We might want to use your phone."
"Yes, surely . . ." The woman gestured at a black telephone next to the cash register and retired to the rear of the shop, where she perched on a high stool and continued to rub her hands.
"Eat her goddamn baklava and your nuts'd probably wind up sealed in a bottle with a genie," Lucas muttered.
Fell glanced back and said, "Shh," but smiled and shook her head. "Fuckin' midwestern white guys, it must be something out there, wall-to-wall Wasps . . . ."
"Look," Lucas said.
Two men in sport coats and slacks were walking up the street, not looking at Lacey's building. One was beefy, the other rail-thin. Their sport coats were too heavy for a New York summer, the kind of coat called "year-round" by the department stores, too hot in summer, not warm enough in winter. The beefy one walked stiffly, as though something were wrong with his back; the thin one showed a cast on his left arm.
"Cops," Fell said. She stood up. "They look like cops."
"The sonofabitch with the cast is the guy who whacked me, I think," Lucas said. Fell took a step toward the door, but Lucas caught her by the arm and said, "Wait, wait, wait . . ." and backed toward the counter and picked up the phone, still watching the two cops. They passed Lacey's building, strolling, talking too animatedly, phony, walked on until they were in front of the next building, then stopped.
Lucas punched Lily's office number into the telephone. She picked it up on the second ring. "I'm at Lacey's place . . . ."
"How'd you get . . . ?"
"I lied. And the Robin Hoods just walked in, we're watching them across the street. So it's O'Dell . . . ."
"Can't be. He hasn't touched a phone."
"What?"
"I'm with him now. In his office."
"Shit . . ."
Across the street, the Robin Hoods had turned and had started back toward Lacey's. One drew a pistol while the other dropped a long-handled sledge from beneath his jacket.
"Get me backup . . ." Lucas said. "Jesus—they're going in. Get me backup _now._ "
Lucas dropped the phone back on the hook. "Let's go," he said. "Get on my arm, really drag on it, like we had a few too many."
They went out the door and Lucas, hat tipped down, wrapped an arm around Fell's shoulder and put his face close to hers. The two cops paused just before they passed the windows in front of Lacey's, looked around one more time, saw Lucas and Fell fifty feet away. Lucas pushed Fell into a building front with one hip, groped at a breast with his free hand. She pushed him away, and the two cops went to the door.
They were running now.
The cop with the hammer stopped, pivoted, swinging his hip like a golfer. Backswing and drive, the hammer flashing overhead.
The hammerhead hit the door just at the handle and it exploded inward, glass breaking, wood splintering.
The cop with the gun and the cast went through; the other dropped the hammer and drew his pistol. Then he went in, crouched, focused, straight ahead.
"Go," said Lucas. His .45 was in his hand, and he was at the door in three seconds. Through the door. The two cops were inside, their pistols pointing up an open stairway, and Lucas dropped in the doorway, screaming, "Police, freeze."
"We're cops, we're cops . . . ." The cop nearest Lucas kept his gun pointed at the stairs.
"Drop the piece, drop, drop it, God damn it, or I'll blow your fuckin' ass off, drop it . . . ."
"We're cops, you asshole . . . ." The heavyset cop was half turned toward him, his gun still pointed up the stairs. The pistol was black with a smooth, plastic look about it, a high-capacity Glock 9mm. This guy wasn't using the issue crap from the department.
"Drop it . . ."
Fell came in behind, her gun out, searching for a target, Lucas feeling the black barrel of the cut-off Colt .38 next to his ear.
"Drop it," Lucas screamed again.
The slat-thin cop, who was closest to the door, dropped his weapon, and Lucas focused on the other, who was still looking uncertainly up the stairs. The disarmed cop said, "Jesus, you asshole, we're plainclothes for Bekker . . . ."
Lucas ignored him, focused on the other gun: "Said drop the fuckin' weapon, jerkweed; you assholes beat the shit out of me, and I'm not in the mood to argue. I'll fuckin' pull the trigger on you right now . . . ."
The cop stooped and laid his gun on the floor, glanced at his partner. "Listen . . ."
"Shut up." Lucas looked at Fell. "Keep your gun up, Bekker's here somewhere."
"Lucas, Jesus . . ." Fell said, but she kept the gun up.
Lucas motioned the two cops to a steam radiator, tossed them a set of handcuffs. "I want to hear them click," he said.
"You motherfucker, I oughta fuckin' pull your face off," the heavy one said.
"I'd kill you if you tried," Lucas said simply. "Cuffs."
"Motherfucker . . ." But the two cuffed themselves to the radiator pipe. Lucas looked up the stairs.
"Now what?" asked Fell.
"Backup's on the way, should be here." He kept the .45 pointed at the chained cops.
"You're fuckin' up," said the thick cop.
"Tell that to O'Dell," Lucas said.
"What?" the cop said. He frowned, puzzled.
Lucas shifted around behind him, his .45 pointed at the guy's ear. "I'm going for your ID, don't fuckin' move . . . ." He slipped his hand inside the cop's coat pocket and came out with a badge case. "Now you," he said to the other one.
When he had both IDs, he stepped back and flipped them open. "Clemson," he said. "A sergeant, and Jeese . . ." Lucas looked at the man with the cast, Clemson, and said, "That's what you yelled—you yelled 'Jeese.' You thought he left you behind, running off like that. I thought you yelled 'Jesus' . . ."
"Here comes the cavalry," Fell said. A blue Plymouth jerked to a stop at the door, and they heard screeching tires from up the block. A uniform came through the door, his gun out.
"Davenport and Fell," Lucas said to them, holding up his badge case. "Working for Kennett with the Bekker team. These guys are cops too, but they're cuffed for a good reason. I want them left like that, okay?"
"What's going on?" the uniform asked. He was a sergeant, older, a little too heavy, uneasy about what he'd walked into. Another car screeched to a stop outside.
"Politics," Lucas said. "Somebody's got their tit in a wringer and the top guys are going to have to sort it out later. But these guys will shoot you if they have a chance. They already shot one cop . . ."
"Bullshit, motherfucker," said one of the cuffed cops.
" . . . So stay cool. Their weapons are on the floor, but I haven't checked them for backup pieces, which they've probably got."
"I don't know . . ."
Two more uniforms squeezed in, their pistols in their hands.
"Look, half the goddamn department will be here in five minutes," Lucas said. "If we're fucked up, we can always apologize to each other later. For now, just freeze the place."
"What about you guys . . . ?" the uniform asked.
"We're going upstairs. You stay here, don't let anybody or anything in or out. Just freeze the scene and be careful. Bekker might be down below, for all we know, and he's armed."
"This is Bekker?"
"This is Bekker," Lucas confirmed. To Fell, he said, "Come on. Let's get him."
## CHAPTER
28
Lily called the patrol lieutenant at the Fifth Precinct and ordered backup squads to Lacey's building: "It's Bekker," she said. "Get them there _now._ "
She dropped the phone back on its cradle and sat down, heavily, in O'Dell's visitor's chair, sorting it out.
_They were in the car_ . . .
O'Dell peered at her across his expansive desk. "What was that all about?" he asked. "The call from Davenport? I believe I was mentioned." His voice was ugly, peremptory. Cold.
Lily shook her head.
"I want to know what he said, Lieutenant," O'Dell barked.
"Shut up, I'm thinking," she said.
O'Dell's eyes narrowed and he sat back. He'd been a politician for five decades, and he instinctively reacted to the warning tone in her voice. Balances had changed somewhere, and he didn't know exactly where. He tried a probe.
"I won't be maneuvered, Lieutenant," he said, emphasizing the rank. "Perhaps a precinct-level job would be more your style after all."
Lily had been peering at the wall above his head, her lips moving slightly. Now she dropped her eyes to his face: "You should have wiped out the ticket requisition for Red Reed before you sent him to South Carolina, John. I've got the ticket vouchers with your signature, I've got the reports on his alleged statements, I've even got the Columbia University transcripts showing that he took classes you lectured at. I also know you fixed at least one drug arrest for him. So don't give me any shit about precinct-level jobs, okay?"
O'Dell nodded and settled in his chair. This could be handled. Everything can be handled by he who waits. He sat silently as she stared at the wall above his head. Finally, a tear trickled down her cheek and she said, "I need your help with the computer."
"What about the Red Reed stuff?" O'Dell asked.
"I'm not going to _use_ it, for Christ's sakes. I mean, I can't conceive of any circumstances that I'd use it. It was just something . . . I found out."
O'Dell grinned in spite of himself. This could be handled, all right. The question now was, who would do the handling? "Davenport," he said. "You _told_ me not to underestimate him. But he looks like a fuckin' brawler with that scar on his face, and what he did to Bekker . . ."
"Two Robin Hoods just showed up at Bekker's hideout. Lucas is going to take them."
"What?" Now O'Dell was confused.
"The computer?"
"Tell me what's going on . . . ."
"I want you to run Copland against Kennett."
O'Dell stared, his thick lips going in and out as he did the calculations, a nursing motion, wet and unpleasant. "Oh, no," he said. He turned, pulled himself across to the computer terminal, flicked a switch, waited until the computer booted up, entered a user name and password, and began the process.
The matching run took ten minutes. A double column of dates and times marched down the screen.
"All so many years ago," O'Dell said tonelessly, reading down the list. "They must've been like father and son. Copland broke him in on the beat. Copland was a tough old bird. He busted more than a few heads in his day."
"Kennett planted him on you. How long ago?"
O'Dell shrugged. "Five years now. He's been driving me for five years. He must have a microphone arrangement in the car, or a bug—or maybe he just pulled out some sound insulation, so he could hear us talk. Every damn thing we said." He looked at Lily. "How?"
"Lucas looked at everything, figured that Robin Hood was either you or Kennett . . . . He trusted my judgment that it wasn't Kennett. At least, he said he trusted my judgment. And he likes Kennett."
"I'm mildly flattered that he thought I could do it," O'Dell said. "So you and Davenport set me up?"
"He suggested that I cover your phones, then plant some information with you and see what happened. Watch where it went. We hadn't agreed on exactly what to do, we were going to talk about it tonight. Then this came up. When he called us with the Bekker thing, he wasn't at Citibank. He was already watching Bekker's place. He expected you to call somebody and maybe send somebody down, some Robin Hoods. And some showed up. But I've been with you . . . ."
O'Dell said, "Now what?"
The tears had started down her face again, but she seemed unaware of them. "What do you mean?"
He made a questioning gesture with his hands, palms up. At the same time, an oddly satisfied expression had settled on his face. "You seem to be running things for the time being. So what do we do?"
She looked at him for a moment, then said, "Call Carter, with Kennett's group."
"Yeah?"
"Tell him what's happening with Bekker, but tell him to cut Kennett out of the loop."
"What about you?"
"Don't ask," she said. She stood and wobbled toward the door of her own office. "Don't fuckin' ask, 'cause I don't know."
## CHAPTER
29
Bekker crouched over Bridget Land, his scalpel in hand, frozen, humming . . . .
When the front door came down, he snapped back, looked down at himself, as though to make sure he was still there, and then at the woman on the table, the scalpel, the monitoring equipment. He heard the footsteps, then the shouts.
Too soon, they'd come too soon, when he was so close.
A tear ran down his cheek. His life had been like this, misunderstood, tormented, unappreciated. Bridget Land, still alive, but hurt, strained away from him, silently . . . .
To do one more would only take minutes, he thought. If he could hold himself together, if they didn't come down too soon.
But Davenport was coming. The gun. He turned, the scalpel in front of his face. The gun was in the other room.
Two impulses fought for control. One propelled him toward the gun, for Davenport; the other told him to finish with Land. Maybe Land would be the transcendent one . . . .
"Don't shoot me in the ass," Lucas said.
He edged up the stairs, Fell two steps behind. Her face was pale, determined, her pistol at Lucas' waist and to the left.
"Just don't roll left," she said.
"Uh-uh . . ."
The smell of marijuana was steeping from the walls, and something else. Lucas sniffed, frowned. Cat urine? And the marijuana odor was years old, not Bekker. In any case, Bekker wasn't much interested in the weed.
At the corner, the first landing, Lucas could see the second-floor door standing partly open, hear Fell breathing below and beside him, smell her faint scent under the odors of the grass and cat piss . . . .
He moved up slowly, across a landing, back against the wall. With the tip of his .45, he pushed the door open. A hall led away, past a closet door, into a living room; he could see the left edge of a television screen. There was no movement, no sound. And the room lacked the peculiar spatial tension of a person in hiding. It _felt_ clear.
"Going in," he whispered.
He stepped past the open doorway to another flight of stairs, the second flight stacked with cardboard cartons, the cartons grimy with years of dust and flaking paint.
"Move," he whispered to Fell. She nodded and eased past him, leveled her gun through the door.
"Go," she whispered back. Lucas crouched, took a breath, then scuttled through the open door on his hands and knees, one hand pushing, his gun extended toward the living room arch, searching for movement, for an anomaly . . . . Nothing.
He stood, held up a hand cautioning her, did a quick head-juke to scan the living room again, then went in. When he was sure it was clear, he waved her in. They checked a sitting room and a dining room; found a pair of glasses lying beside the couch, thick lenses, bifocals. Old-lady glasses. Checked the closets, groped through them. Nothing.
The kitchen was small, smelled of boiled beets, boiled cabbage, boiled carrots, porridge. A pool of water shimmered below the refrigerator. Fell squatted next to it, then looked up at the refrigerator. The main door wasn't quite closed, and water dripped from the bottom of it. She pointed, then put her finger to her lips.
Lucas, standing beside her, reached out, took the door handle. Nodded. Jerked it open.
"Aw, shit," Fell said, lurching away from the refrigerator.
Mrs. Lacey hadn't fit that well, but Bekker had managed to crush her into the limited space. Her head lay at right angles across her shoulders, and the light behind her head glowed like a perverse advertisement. Her eyes were bloody holes. A dozen cans of Coke were carefully stacked around her body, one jammed between her twisted arms and her chest. Two dead cats were stuffed in a plastic meat compartment, their tails trailing out.
"Jesus. Jesus." Lucas backed away. "Let's go up the next one, but make it quick."
"You think he's up there?" Fell asked doubtfully. She was staring at the refrigerator, her throat working.
"No. If he's in the building, he's down—I don't feel anything up here."
"Air's too quiet," Fell said. "C'mon, you cover me . . . ."
She went ahead for the next flight, climbing past the cartons, through the dust. At the top, they found three bedrooms and an old-fashioned bath. They checked the closets, the shower, under the beds. Nobody home.
"Down," Lucas said.
"How about the roof?"
"We'll send a couple of guys up—but Bekker would look for a hole, not a perch."
Six cops were spread through the first floor, all looking up apprehensively when Lucas and Fell hurried down the stairs.
"He killed an old woman and stuffed her body in the refrigerator," Lucas told the patrol sergeant, flicking a thumb at the stairs. The two Robin Hoods watched silently from the radiator, their hands still looped through the cuffs. "We went through both floors, nobody home. Send a couple of good people up, see if they can find the roof access. We didn't check that. Tell them to be careful. He's got a gun."
"I'll go myself . . . ."
"No. You stay here. You've got enough rank to keep these assholes cuffed up," Lucas said, nodding at Clemson and Jeese. "There'll be more people coming soon, just hang on. We're gonna do the basement . . . ."
"Take it easy, then," the sergeant said, still uneasy, looking at the two sullen cops chained to the radiator.
The stairs were clean; they looked used. Lucas edged down, taking it easy, leading with the .45, while Fell crouched at the top, focused on the corner at the bottom. If Bekker came around, she would see him before Lucas. But as Lucas reached the corner, her firing line was cut off and he held up a hand to caution her.
Crouching on the bottom step, he did a head-juke around the corner, a one-eyed peek at waist level. A short concrete-floored hallway ended at a green wooden door. A bare bulb hung in the hall above the door. He groped around the corner for the switch, found it, flicked it on.
He stood and crooked two fingers at Fell and she padded down the stairs. "Get that sledgehammer and bring back somebody who knows how to throw it."
Fell nodded. "Be right back," she whispered.
Lucas waited by the door, the gun pointed at the knob. If Bekker was in the basement, and alive, he'd know the cops had arrived. But if he was waiting with a gun, it was critical that he not know the instant that the door would come down . . . .
Fell came back down the stairs with the sergeant and the sledge.
"We got an entry team coming," the sergeant whispered urgently. "They got the armor . . . ."
Lucas shook his head. "Fuck it. I'm taking him . . . ."
"Listen, these guys can take him, no problem . . . ."
"I'm going," Lucas said. He looked at Fell. "What about you?"
"I'll cover, or go in, whatever . . . ."
"God damn it, you're gonna get our asses shot," the sergeant whispered.
"Give me the sledge," Lucas said.
"Listen to me."
"Give me the fuckin' sledge . . . ."
"Ahhh, shit . . ." The sergeant shook his head and hefted the hammer. "I'll swing it, you assholes back me up. I'm going to hit that fucker once, and then I'm on the floor."
"Let's do it," said Fell.
Bekker wandered through the murky basement, trying to remember why he was going to the couch. A song went through his head:
_Jesus loves me this I know, for the Bible tells me so._ . . .
Sung at a funeral, sometime, way back, he could remember a bronze coffin that sat higher than his head and the choir singing. It was all very sharp, as though he'd just stepped into the picture . . . .
A spider brushed his cheek, tickling, and Bekker snapped out of the funeral picture. Something thumped overhead. That was it. The noise. He had to go to the couch because of the noise overhead.
The couch had been pushed out from the wall, and he stepped behind it and sat down on the rug. The gun was waiting, cheap chrome steel. Loaded. Two shots. He picked it up. Said, Hello, put it in his mouth, sat, like a man with his pipe, then took it out and looked down the barrel.
_Hello_ . . .
His finger tightened, he felt the pressure of the trigger, took up the slack . . . and his mind cleared. Clear as a lake. He saw himself, huddled in the corner of the basement. Saw Davenport come in. Saw himself, hands crossed over his chest, shoulders pulled in, head down.
Saw Davenport coming closer, screaming at him; saw himself rocking back and forth on his heels. Felt the pistol in the bottom hand on his chest, concealed. Saw Davenport reaching out to him, ordering him to turn; Davenport unaware, unknowing, unthinking. Saw himself reach out with the derringer, press it to Davenport's heart, and the explosion and Davenport's face . . .
The sergeant looked at Lucas, raised an eyebrow. Ready? Lucas nodded. The sergeant took a breath, raised the hammer overhead, paused, then brought it crashing down. The door flew inward, and the sergeant hit the ground. There was no immediate fire from the dark room, and he scrambled back past Fell to the stairs, groping for his gun.
"Too fuckin' old for this shit," he said.
Lucas, focused on the room, said, "Flashlights."
"What?"
"Get some flashlights . . . ."
With quick peeks around the corner, they established that the interior of the basement wasn't quite dark. A light was on somewhere, but seemed to be partially blocked, as though the thin illumination were seeping through a crack in the door, or coming from a child's night-light. Lucas and Fell, looking over the sights of their weapons, could see the blocky shapes of furniture, a rectangle that might be a bookcase.
"Got 'em," the sergeant said.
"Poke them around the corner, hit the interior, about head high. Keep your hand back if you can. Tell me when you're going, I'll shoot at a muzzle flash," Lucas said. He looked at Fell, saw that she was sweating, and grinned at her. "Life in the big city."
The cop nodded. "Ready?"
"Anytime."
"Now."
The cop thrust the light around the corner, and Lucas, four feet below, followed with the muzzle of his gun, and his arm, and one eye. No movement. The sergeant leaned a bit into the hallway, played the light around the interior.
"I'm going," said Lucas.
"Go," said Fell.
Lucas scrambled across the floor to the apartment door, then, flat on the floor, eased his head and shoulders through the door, reached up, flicked a light switch. A single bulb came on. Nothing moving. He crouched, and Fell eased down the hall.
"What's that?" she whispered.
Lucas listened.
_Jesus loves me_ . . .
Not a child's voice. But not an adult's, either—nothing human, he thought. Something from a movie, a special effect, weird, chilling.
_For the Bible tells me so._ . . .
"Bekker," Lucas whispered. "Over there, I think . . ."
He was inside the apartment, duckwalking, the .45 in a double-handed grip, following his eye-track around the apartment. Fell, behind him, said, "Covered to the right."
"I got the right, you watch that dark door . . . ." The sergeant's voice. Lucas glanced back, quickly, saw the older man easing inside with his piece-of-shit .38.
"Got it," Fell agreed.
"He's in the corner," Lucas said. He half stood, looking at a velour couch. The couch was pushed away from the wall, and the unearthly voice was coming from behind it.
"Bekker," he called.
_Jesus loves me_ . . .
"Stand up, Bekker . . . ."
_This I know_ . . .
Lucas focused on the couch, crept up on it, the gun fully extended. Up close, he could see the top of Bekker's head, shaven, smooth, bobbing up and down with the simple rhythms of the song.
"Up, motherfucker," he yelled. And to Fell and the cop: "He's here, got him . . ."
"Watch a gun, watch a gun . . ."
Lucas, pointing his weapon at the top of Bekker's head, slid around the side of the couch and looked down at him. Bekker looked up, then stood, hands across his chest, rocking, humming . . . .
"Turn around," Lucas shouted.
Fell moved up beside him . . . .
"Nuttier 'n shit," she whispered.
"Watch him, watch him . . ."
She stepped around to get a better angle, then batted at her face and batted again, then waved her hand overhead.
Lucas, glancing sideways: "What?"
"I'm tangled . . ."
Bekker's head turned, like a ball bearing rotating in a socket. "Spiders . . ." he said.
The sergeant, near the kitchen door, coming up slowly, punched a light switch, and Fell groaned, weakly, thrashing at the objects that hung around her head.
"Get away," she choked. "Get away from me . . . ."
They hung on individual black threads from a bundle of crossed wire coat hangers, floating in their separate orbits around Fell's head, wrinkled now, drying, the varicolored lashes as sleek as the day the eyelids were cut from their owners . . . .
Fell staggered away from them, appalled, her mouth open.
"Get him," Lucas said, his pistol three feet from Bekker's vacant eyes. The sergeant took a step forward. Behind Fell, a thin shaft of light cut through a crack in a door. The light was hard, sharp, blue, professional. As the sergeant stepped forward, Fell pushed the door open.
Bekker took a step toward Lucas, his hands crossed on his chest. "Spi . . ."
An old woman lay there, bound and wired silent, her eyes permanently open now, staring, white eyeballs, the skin removed from her chest . . . .
Alive . . .
"Aw, fuck," Fell screamed. She pivoted, the gun coming up, her mouth open, working, her hands clutching.
Lucas had time to say, "No."
Bekker said, " . . . ders." And one hand dropped and the other swung up, a glint of steel. He thrust the derringer at Lucas' chest . . .
. . . and Fell fired a single .357 round through the bridge of Michael Bekker's nose and blew out the back of Michael Bekker's sleek, shaven head.
## CHAPTER
30
The walls of Lily's office seemed to melt, and Petty was there, the adult face superimposed on the child's face, both of them together.
And then Kennett's face.
Kennett's face in the dark, in Lily's bedroom. Must've been in winter: she'd bought a Christmas tree, shipped into a lot on Sixth Avenue from somewhere in Maine, and she could remember the scent of pine needles in the apartment as they talked.
No sex, just sleeping together. Kennett laughing about it, but unhappy, too. His heart attack not that far past . . .
"Hanging out with a geek," he said. "I can't believe it. I'm not enough, she's got a geek on the side."
"Not a geek," she said.
"All right. A dork. A nerd. Revenge of the Nerds, visited on Richard X. Kennett personally. A nerd may be dorking my woman. Or wait, maybe it's a dork is nerding my woman. Or wait . . ."
"Shut up," she said, mock-severely. "Or I will fondle your delicate parts and then leave you hanging—in good health, of course."
"Lily . . ." A change of tone. Sex on the mind.
"No. I'm sorry I said it. Kennett . . ."
"All right. Back to the dork . . ."
"He's not a dork. He's really a nice guy, and if he cracks this thing, he could go somewhere . . . ."
She'd talked, Lily had, about the Robin Hood case. She'd talked in bed. She'd talked about the intelligence guys who'd stumbled over it, she'd talked about Petty being assigned to it, she'd talked about computers.
Not all at once. Not formally. But bits and pieces. Pillow talk. But Kennett got most of it. With what Copland overheard, and what Kennett got in bed, they must've known it all.
Petty's image floated in her mind's eye, his hair slicked down, his red ears sticking out, running down the Brooklyn sidewalk with the paper overhead, so happy to see her . . . .
"I killed you," she said to his image, speaking aloud. Her voice was stark as a winter crow. "I killed you, Walt."
## CHAPTER
31
The river was black as ink, but thick, oily, roiled, as it pushed the last few miles toward the sea. A full moon had come up in the east, red, huge, shrouded by smog over the city. Lily waited until the elderly night guard and his dog were at the far end of the marina, then used her key at the member's gate.
The docks were cluttered, as always, badly lit by widely spaced yellow bug lights. Out in the water, anchor lights shone off the masts of a half-dozen anchored boats. Here and there, lights showed at portholes, and a light breeze banged halyards against aluminum masts, a pleasant whipping tinkle like wind chimes. The smell of marijuana hovered around a small Capri daysailer and a man was giggling inside the tiny cabin. She walked out of the marijuana stink into the river smell, compounded of mud and decaying fish.
"Lily." Kennett's voice came out of the dark as she approached the _Lestrade._ He was sitting behind the wheel, smoking a cigarette. "I was wondering if you'd come."
"You know about Bekker?"
"Yeah. And that I've been cut out of the loop."
Lily stepped into the cockpit, sat down, staring at him. His face was flat, solemn; he was looking steadily back. "You're Robin Hood," she said.
"Robin Hood, bullshit," he said wearily. He flicked the cigarette into the water.
"I'm not wearing a wire," she said.
"Stand up, turn around." She stood up and Kennett ran his hands down her, between her legs. "Gimme the purse."
He opened the purse, clicked on an electric light that hung from the backstay, looked inside. After poking inside, he took the .45 out of its holder, dropped the magazine and shucked the shells out into the water. Then he jacked the slide, to eject the shell in the chamber. The chamber was empty, and he shook his head. "You oughta carry one under the hammer."
"I'm not here to talk about guns," she said. "I'm here to talk about you being Robin Hood. About using me as a dummy to spy on O'Dell. About killing Walt Petty."
"I didn't use you as a dummy," he said flatly. "I got with you because I liked you and I'm falling in love with you. You're beautiful and you're smart and you're a cop, and there aren't many women around I can talk to."
"I don't doubt that you like me," she said, squaring off with him. "But that didn't keep you from running me. On the way up here, I was remembering when we'd lie down below there, in the berth, and you running those goddamn fantasies about what O'Dell did for sex. Do you remember that? You must've scripted those things, to get me talking about O'Dell. And before that, talking about Walt. When I think of the things I told you, because I felt secure. Because you were a lover and a brother cop. Jesus Christ, every time we got into bed, you were pumping me for information."
"Christ, Lily . . . Lily, if you told me anything about O'Dell or Petty . . . it was by-product. I wasn't sleeping with you to get information. Jesus, Lily . . ."
"Shut up," Lily said. She reached overhead and pulled the chain on the backstay light and they were plunged into the dark again. "I want to know some shit. We've got Jeese and Clemson, Davenport got them, and we know about Copland . . . ."
"I knew Davenport was dangerous," Kennett said quietly. "I really didn't underestimate him. I knew he was a _really_ dangerous sonofabitch when he looked up Gauguin, about the necktie. And I couldn't help liking him."
"Is that why your guys tried to beat him up, instead of just whacking him?"
Kennett grinned: she could see his teeth. Not a happy smile, a rueful one. "Another mistake," he said. "You start feeling that everything in New York is _more._ That a small-town guy could never hold off a couple of real New York pros. So we were just gonna break a few ribs, maybe. Something that'd take him off the street for a month. They said he was quick as a pro fighter. They were pissed, said that if they'd been a half-inch slower, he'd of blown them up, he'd of had his .45 out . . . ."
"They were lucky," Lily said. "Why didn't you try again?"
Kennett shrugged. "At that point, we figured it was either kill him or forget him. He didn't seem . . . close enough . . . to kill. And I don't know if the guys would've done it anyway. Petty was already hard to stomach. Davenport's message to O'Dell, the one Copland picked up. That was fake?"
"Not completely. It was Davenport who found Bekker, all right. He was feeding the message to O'Dell to see if any hitters showed up. They did, but I was with O'Dell the whole time. He didn't make any calls. So I started thinking about it."
"God damn it. I thought about skipping Bekker."
"You should have."
"Couldn't. Didn't know what he'd say about . . ." He stopped, remembering.
"About the guys he saw hit Walt. Jeese and Clemson. Thick and Thin."
"No," Kennett said evenly. "It wasn't them."
"Bullshit," she flared. "They fit."
"No. It wasn't."
"Who, then?"
"I won't tell you, but Jeese and Clemson, no." He pulled at his lip. "Old Copland. A good guy. What happens to him?"
"O'Dell will think of something . . . . How many of you are there? And how many people have you done?"
Kennett shook his head. "There are . . . several. Some singles, some two-man teams. None of them knows the others, and I won't tell you who they are."
"We can put Jeese and Clemson in Attica if we want—assault on a police officer with a firearm. And if O'Dell wants to fix it, I'm sure we can find a problem with Copland's pension. He'll spend his last twenty years sitting on a park bench. Or rolled in an army blanket on a sidewalk."
"Don't fuckin' do that," Kennett whispered.
"That's what happens when you lose," Lily said, her voice like ice.
"We were doing right," Kennett said. "I'll call it off. Walk away, and I'll call it off. I'll quit the force, if you want."
"What, so you can write for the _Times_? You'd be a bigger danger there than where you are now," Lily said.
"So what do you want from me?"
"I want the goddamned names."
Kennett shook his head. "No. Never happen. If I gave you the names, only two things could happen: a lot of good guys would get ripped off, or O'Dell would set up his _own_ little force of stormtroopers. I'm not going to let any fat, puling, alcoholic fixer do that, I won't . . . ." His voice grew cold as he said it. He bared his teeth and added, "I really like you. But the worst thing you do is, the worst thing about you, is that you associate with that . . . that . . . cunt O'Dell."
"I'm the cunt," Lily said. "I'm the one you rolled for information."
"Fuck you, then," Kennett said, and turned away. "You want to make something out of it, make it in court. I'll tear you up. Now take your ass off my boat."
"I've got another question before I go."
"What?"
"Why Walt?"
Kennett stared at her a moment, then dug in his shirt, found a pack of cigarettes, shook one out, lit it with a match. Tossed the match overboard: they heard it hit, the hiss hanging in the damp air.
"Had to," he said. "Him and his fucking computers. When I started this, nobody really knew about computers and what they could do. They were like electric filing cabinets. Looking in a computer was like snooping through papers on somebody's desk. We didn't know that every time we went into a file, we left tracks. Petty nailed us down. We had to have time to get into the machines, to fix things. We did that. The information's gone now." He looked downriver, at the Manhattan glittering along the river, the arcs of the bridges. "Listen, Lily. If you could take five hundred or a thousand people out of Manhattan, you could make it eighty percent safer. You could make it a paradise."
"Not a thousand," she said. "Maybe ten thousand."
"No. No, not really. A thousand would do it. We couldn't take down a thousand people, probably, but we could make a difference. Arvin Davies. You look at him? Was he one of the people . . ."
"Yes."
"We think . . . intelligence estimates . . . that he committed up to a hundred crimes, all sorts: assaults, burglaries, rapes, murder. He could have done a hundred more. Now he won't."
"You can't make that decision."
"Sure I can. And somebody has to," Kennett said, looking at her. "Your average junkie does fifty or a hundred burglaries for every time he gets caught, and for small burglaries, chances are he'll be right back out on the street. Plea-bargains out, or he'll do thirty days or six months or something. Not enough. If we let all the onetime passion killers out of prison and put all the junkies inside, Manhattan would be a garden spot. Even the ones we took off . . . Christ, we knocked down a thousand violent crimes a year, just the ones we took down."
"How many were there?"
He shook his head. "You don't need to know. But that's why."
"That's why you shot Petty? So we'd have a garden spot?"
Kennett turned away. "We didn't like doing that. But we had no choice . . . . O'Dell is trying to frame me, by the way. Supposedly had a witness who saw me when Waites was gunned down."
"I know."
His eyebrows went up. "You know?"
"Davenport found the kid who supposedly saw you. Found him in Charleston and broke him down. He knows it was phony."
Kennett smiled. "When he went to Minneapolis, he went to Charleston the next day. I thought it was weird that he took the day off—weird for a guy like Davenport."
"How about the others? Waites was a loudmouth, but . . ."
"They nurtured it, the festering. My God, look over there, look at that city, think what it could be . . . ."
She looked across the water at the twinkling lights, like the lights of the Milky Way, seen large. "And you sold it out. And used me like a fucking Kleenex."
"Bullshit," he said. His face was getting red.
"When Walt was killed, I came over here and cried on your shoulder, and you took care of all the arrangements and patted me on the head and took me down below and made love to me, comforting me. I can't believe I did it."
"Yeah, well . . ."
"Well, what?"
"That's life." His teeth were clenched. "Now, go on, Lily, get the hell out of here."
Lily stood, took a step toward the dock. Then another step, toward Kennett.
"What . . ." Kennett began.
She hit him, open-handed, hard: a slap that almost knocked him down. He took a step toward her, hand on his face, and caught her arm. "Lily, dammit!"
"Let go of me," she said. She tried to pull away, but he held on, and for a moment, they struggled together, his face getting redder; then suddenly, he pinched his shoulders and let his hand drop away.
He turned, seeming to crouch, then went to his knees. "Oh, Jesus," he gasped. "Lily . . . in my bag, down below . . ."
His pills. His pills were in the bag. She started to turn toward the cabin.
A spasm hit him and he went flat in the cockpit, his face straining, the tendons standing out in his neck. "Lily . . ."
She stopped. Looked at the cabin and then back at him. And then carefully, as if in slow motion, she climbed out of the boat, stood on the dock a second, looked at the city and then back down at Kennett. His face was chalky, his mouth open, straining, his eyes large and staring. His hand scrabbled along the deck, as though he were trying to get hold of it. "Lily . . ."
"Say hello to Bekker," she said.
## CHAPTER
32
O'Dell sat in his semidarkened office, an air of satisfaction about him, like a bullfrog who'd snapped up a particularly tasty fly. "I really don't give a fuck what you think," he told Lucas.
"Which makes me want to come across the desk and slap the shit out of you," Lucas snarled.
"The New York jails aren't pretty," O'Dell said, mildly. "I could guarantee you a tour . . . ."
Lucas shook his head. "Nah. You wouldn't do that. I spent too much time with Red Reed. We had witnesses. So I slap the shit out of you, you put me in jail, and I tell the papers about Reed, and tell them that you hid a key witness in the murder of a well-known black politician. You'd be right in there with me."
O'Dell seemed to think about it for a minute, then sighed and half closed his heavy-lidded eyes. "All right. But look, if you're gonna slap the shit out of me, why don't we get it over with? I need some sleep."
They sat quietly for a minute, then Lucas said, "You know I won't. But you owe me, God damn it. You got me whacked by Kennett's hoods. What I want to know is, how much was set up? Did you know it was Kennett? Is Lily in it? How about Fell? And who else?"
"Lily's okay—she never had anything to do with it. And Lily says you believe Fell was an alarm. I don't know if I believe it, but I can see the possibility . . . ."
"Kennett?"
"Yeah, I knew about Kennett and a couple more—and frankly, you and Lily should have known that," O'Dell said. "Petty's investigation wasn't a TV show. He didn't sneak off and do all the work and keep all of his conclusions to himself. He came up and sat here every day and told me what he thought. We had Kennett and a couple more people spotted—not Copland, unfortunately. We _didn't_ know that Kennett had his own computer people. We figured we could go into the system anytime, print out our evidence. Then Petty got killed and his printouts were lifted. When I went back into the system, the files had been trashed. All I had were a few names and no way to push."
"So you set us up."
O'Dell smiled, still pleased with himself. "Yes. Lily had talked about you. Said you were smart. And I saw one of your simulations. So I put Kennett on Bekker, and you on Kennett, and brought Fell to work with you, and had Lily running you on the side. With all that pressure, something had to blow. Anyway, I had nothing to lose."
Lucas thought about it, stood, stretched, yawned, wandered to O'Dell's window, pulled back the heavy plush drapes and looked out at the twinkling city. "This goddamn place is one big patch, you know? Have I given you my rap on how the place is one big patch?"
"Yeah."
"And I was another one."
"Yeah."
Lucas stretched again, then wandered across the room toward the door. "Nice game," he said.
O'Dell looked at him, then laughed, low and long, genuinely delighted. "It was, wasn't it?"
Lucas sat behind a round, simulated-wood table the size of a manhole cover, in a plastic bar full of plastic pictures of old airplanes. Through the clear Plexiglas walls, he could watch the people streaming out toward the departure gates. He glanced at his watch: three twenty-seven in the afternoon, more or less. With a Rolex, he'd discovered, more or less had to be good enough. He sipped at his Budweiser, not interested, just holding his seat.
Fell showed up at three-thirty, thin, bird-gawky, tough. And maybe angry or something else. She stopped near the end of a long queue for the security gates, looked both ways, and spotted the bar. She paused again at the door, and Lucas raised a hand. She saw him and threaded her way through the tables. When she saw his suitcase by his leg, she looked from the case to Lucas and said, "So I was a three-night stand, or whatever it was."
"Not exactly," Lucas said. "Sit down."
She didn't sit down. Instead she said, "I thought we might go someplace for a while." Tears rimmed her eyes.
"Sit down," Lucas said.
"You fuck," she said, but she sat down, dropping heavily into the chair across from him, hands dangling dispiritedly between her legs. "You said we . . ."
"I thought about asking you to come down to the Islands with me," Lucas said. "I even called out to Kennedy, out to United, to find out what islands we could go to."
She looked down at the tabletop. "Tell me," she said.
"Well, I . . . couldn't." He dug in his pocket and tossed a red matchbook on the table in front of her. The matchbook had a horsehead on it. She picked it up and put it in her purse.
"So you were in the restaurant where Walter Petty got killed," he said. "You told me you weren't."
"Yeah?"
"Yeah. I saw the matches in your apartment."
"When?"
"Well, when we were up there . . . ."
"Bullshit, I got rid of them. When I thought you might be coming over, I saw them, and I thought, 'I got to get rid of these.' I threw them out. So when did you see them?"
He looked levelly across the table at her. "The first day we worked together, I copped your purse, made molds of your keys. The next day I went in."
"You sonofabitch," she said. Then a realization came to her eyes. "You're wearing a wire?"
"No. I like you too much. But the thing is, I can't trust you. Not completely. I thought about going down to the Islands with you and decided I couldn't. I'd eventually talk to you about this, and then . . ." He let the thought dangle, and so did she. He went on: "I tried to think up a lie that would get me back to Minnesota. But I couldn't think of one. And I wanted to tell you why."
"Well. I appreciate it. But you'd have been safe enough. A matchbook is pretty thin . . ."
"There was more than a matchbook. This whole goddamned episode was a game set up by O'Dell. It was so beautiful it makes me laugh. He used every one of us. But anyway—he did a computer run on the victims. You come up way too often. That was a big piece."
She frowned. "Will they get me?"
"No, I don't think so. They think you're an alarm." He explained, and she listened quietly, staring at the floor.
"And you won't tell them different?" she asked, when he finished.
"No. I'm the one who sold them the alarm idea."
"Why?"
He shrugged. "You're a friend."
She looked him over for a moment and then nodded. "Okay."
"If Lily ever found out, though, she very well might kill you. That's another reason I wanted to talk . . . ."
"Did she kill Kennett?" Fell blurted.
"Kennett? No, no, she was downtown with O'Dell all evening."
"Goddamn," Fell said, gnawing a thumbnail. "When I shot Bekker . . ."
"Bekker knew you," Lucas said. "And that's why, in his letter, he wouldn't say anything about Thin. He didn't want people thinking about women killers . . . ."
"Yeah," Fell said. "But that's not why I shot him. I shot him because of their eyelashes, and that woman . . . and everything."
"I know. I mean, I believe it. But why Petty?"
"I didn't want to do Petty," Fell said, voice low, out of gas. "I was there, but I tried to stop it."
"You didn't have to be there . . . ."
"Well . . . I was. If I'd had a couple of more minutes, I think I would've talked . . . the other guy out of it. But Petty came through the door a minute too soon. A minute later and nothing would've happened. At least, not then. Petty had something on us . . . . I'll burn in hell for Petty."
"I doubt it," Lucas said wryly.
"Well, so do I," she said. Then: "I would've liked the Islands, though. Going down with you."
"Yeah, it would have been nice. But I'm the only one who knows about you. You're quick with that gun . . . and you might start thinking about it, if I'm there, laying around."
"I wouldn't," she said, but she couldn't suppress a small grin. "It's interesting that you're scared of me, though."
"Yeah, well . . ."
She sighed. "Fucking trouser-snake cops. So goddamned treacherous."
"And I wanted to tell you about Lily," he said.
"What?"
"She's got a line on a half-dozen of Kennett's shooters. She's gonna be tough, one way or another. But I want you to know two things: they've got no proof of anything. They just want it to stop."
"What's the other thing?"
"The other thing is, if anybody takes Lily, I'll be coming back to town," he said. He'd been watching her, and his eyes had gone hard as granite.
"You oughta be one of us," she said.
"Pass the word on," he said.
"I don't know anybody, except my . . . pal . . . and one other guy. But I'll tell them. Maybe they know more. We don't talk about it. That was one of Kennett's rules. Nobody talks about nothin', he'd say."
"Good rule," Lucas said. He looked at his watch again. "Lily's coming pretty soon."
"Here?"
"Yeah, I've got to talk to her too."
"Then I better get going," Fell said, picking up her purse. She stood and stepped away from the table, then turned back. "Remember when you said something like, 'This place is the armpit of the universe,' the first day we were together?"
"Yeah?"
"Kennett's people . . . we were just trying to make it something else."
"Okay."
"Were we wrong?"
He thought about it for a while. "I don't know," he said finally.
Fell went away and Lucas stared at his beer bottle, making wet O's on the table. After the shooting in the basement, after the dictated statements and interrogations, after the press conference, he'd gone back to the team office. Most of the office staff had gone, but he'd found a computer adept, and said that he needed to look up some information on a couple of cops: Jeese and Clemson.
The computer operator had put him at a vacant terminal, showed him how to call up the files. He'd done it, read through them quickly, then punched in Fell's name. When he'd gotten the file, he'd scanned through to the bottom, found the next of kin: Roy Fell, at an address in Brooklyn. He'd punched in Roy Fell. A file had come up. _Retired,_ it said. Then: _Retrieve Retired File? (Y/N)._
Lucas had pushed the Y key. A photoscan was a simple matter of selecting the right option on a short menu, and Fell's father's face had come up. Heavy face, gray hair, gray mustache, a smile that looked almost painful. Six feet, two inches tall. Born 1930. Bekker had had him pegged almost exactly.
"Thick," Lucas had said aloud.
The computer operator said, "What?"
"Nothing," said Lucas, and he'd shut the terminal down.
Sitting at the airport now, drawing circles with the bottom of his beer bottle, Lucas thought, _You can't walk away from family._ . . .
Lily arrived ten minutes later. Like Fell, she stopped by the security queue, looking for the bar. She saw him as she came in, her face ashen, tired, but controlled.
"You talked to O'Dell," she said as she sat down.
"Yeah."
"He fixed the whole thing."
"Yup."
"When did you know?" she asked.
"In Charleston. I suspected before that—everybody was too close together, everything was too convenient. But I didn't know for sure that he wasn't Robin Hood."
"Do you still think Fell was an alarm?"
"Yeah, I'm pretty sure. Not positive. But I think she was simply set up by Kennett. I mean, she _took_ those Robin Hoods at Bekker's place. She didn't have to: her piece was right in my ear."
"The word is going around that Robin Hood _did_ get Bekker."
"What'd you expect? He got shot to death."
Lily sat for a moment, staring at the fake grain on the tabletop. "When did you know about Dick?" she asked.
"O'Dell tried to set him up—that thing about a white-haired guy killing the politician. I didn't know it was a setup, so even then, I was thinking about him."
"But when . . . ?"
"When we went to Petty's apartment and that Logan woman said whoever came to Petty's apartment seemed to stop before he got to the elevator, and after he got off the elevator, and to take a long time getting to the door . . . ."
"Sure," she said, avoiding his eyes. "Dick."
"Yeah, but I couldn't figure it. I assumed he couldn't drive—that's what everybody assumed—and saw a driver dropping him off at Midtown South. And if he couldn't drive, it wasn't him. If he'd been driven, by Copland or one of his other buddies, he wouldn't have had to walk up all those steps himself. He could have sent the driver in for the stuff. So that pushed me off him for a while. Until the day on the river and you told me that he _could_ drive. That he sometimes drove the four-by-four, and it pissed you off . . ."
"So," she said. "I not only betrayed Petty, I betrayed Dick."
"Ah, come on, Lily, stop sniveling. You were doing the best you could in a goddamned rat's nest," Lucas said.
"And everybody winds up dead," she said.
"Hey." There wasn't much else to say. Lucas looked at his Rolex. "I gotta go. They're probably boarding the plane now," he said.
At the end of the security queue, Lucas faced her, hands in his pockets, and said, "If this was a movie, there'd be a big hot kiss right here and everything'd be all right."
She had eyes that Rembrandt would have painted. "But there's never anything after a movie," she said. "It ends with a hot kiss and you never see the going-back-to-work part."
"The getting-to-be-important part . . ."
"Yeah. And to tell you the truth, if there was going to be a big hot kiss, I thought Fell'd be getting it. I thought you'd be going out to the Islands with her."
"Nah. She's New York, I'm not. Besides . . ."
"What?"
"There really aren't any Islands, are there?"
She looked away from him, thinking of Petty and Kennett. "No," she said after a minute, "I guess not."
There was another moment, and she stuck out her hand.
"Give me a fuckin' break, Rothenburg," Lucas said, and leaned into her and kissed her on the lips, almost, but not quite, chastely. He turned and started through the security check. "If you get another Bekker, give me a whistle. You know . . . ?"
"Yeah, yeah. Jesus," she said, not quite believing him. A tiny smile crinkled the corner of her mouth. "I do know how to whistle."
• • •
[For a complete list of this author's books click here or visit
www.penguin.com/sandfordchecklist](http://www.penguin.com/sandfordchecklist?CMP=OTC-SANFORDCHKLST)
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 3,689
|
# Cover
> The brave and intelligent expect to leave to their posterity the splendour of their public services, embodied in rank and honours. A country that prohibits such a legacy destroys one of the chief sources of its greatness, and blasts the vital principle of public virtue.
>
> — A New Trend on an Old Subject, Dallas, 1791
To my friends,
Joyce Bryant and Bruce Beatty,
builders of what is herein
Her Majesty Queen Elizabeth II,
Queen of Canada
# Table of Contents
1. Foreword
2. Acknowledgements
3. Introduction to the Second Edition
4. Part I
5. Chapter 1
6. Chapter 2
7. Chapter 3
8. Chapter 4
9. Chapter 5
10. Chapter 6
11. Chapter 7
12. Chapter 8
13. Chapter 9
14. Chapter 10
15. Part II
16. Chapter 11
17. Chapter 12
18. Chapter 13
19. Chapter 14
20. Chapter 15
21. Chapter 16
22. Chapter 17
23. Chapter 18
24. Chapter 19
25. Chapter 20
26. Chapter 21
27. Chapter 22
28. Chapter 23
29. Chapter 24
30. Chapter 25
31. Chapter 26
32. Chapter 27
33. Chapter 28
34. Chapter 29
35. Chapter 30
36. Chapter 31
37. Chapter 32
38. Chapter 33
39. Chapter 34
40. Part III
41. Chapter 35
42. Chapter 36
43. Chapter 37
44. Chapter 38
45. Chapter 39
46. Chapter 40
47. Appendix I
48. Appendix II
49. Appendix III
50. Ribbons
51. Notes
52. Bibliography
53. Photo Credits
# Foreword
A nation's custom of conferring honours on its citizens — to reward or to recognize service or merit — has existed for hundreds if not thousands oof years. Regarded by some as unnecessary and by others as an important symbol of nationhood, a national honours system says much about the nation itself, including its approach to egalitarianism and merit and its belief in the importance of recognizing performance, professionalism, excellence, courage, and sacrifice.
In 2005, Christopher McCreery published The Canadian Honours System, and in the foreword to that book I noted that it was: "an important and comprehensive account that should be of interest to all Canadians, as well as a valuable addition to Canadian historical literature. On a subject that is still developing, it will surely need updating from time to time."
That time has come, given that much has happened in the Canadian honours system during the past eight years, and in this second edition, McCreery brings the original up to date. Fifty percent longer than the first, this volume includes a section on the Canadian use of British honours up until 1967 when Canada introduced its own system. Previously, and for more than a century, Commonwealth honours were Canada's honours, something worth remembering and of which many Canadians will be unaware.
This second edition also includes descriptions of new Canadian honours such as the Sacrifice Medal, General Campaign Star, General Service Medal, and Operational Service Medal. Included also are names of Canadians recently admitted to the rare Order of the Companions of Honour and to the even rarer Order of Merit. Included, as well, is the 2012 Queen's Diamond Jubilee Medal, a commemorative medal awarded to considerably more Canadians than in the past, fittingly so perhaps, given that a monarch's celebration of sixty years on the throne is a rare event, indeed.
This book, like its predecessor, remains an important and comprehensive account that should be of interest to all Canadians, as well as a valuable addition to Canadian historical literature. As it demonstrates, national honours and awards are constantly developing to meet the nation's changing needs, and no doubt it, too, will need updating again sometime. But for now, the fact that it is both authoritatively and entertainingly written should ensure that it has a wide readership.
General John de Chastelain, CC, CMM, CD, CH
Ottawa
# Acknowledgements
Editing and expanding upon a pre-existing work is no easy task, and I am indebted to an extended cadre of friends and colleagues who encouraged me throughout this process. First, thanks are owed to Dundurn Press and its presi-dent and publisher, Kirk Howard, for continuing to show confidence in my writing.
Two friends in particular were central to the development of this edition. Major Carl Gauthier from the Department of National Defence's Directorate of Honours and Recognition provided regular encouragement, editorial comments, and overall insight into many elements of the modern Canadian honours system. The deputy chief herald of Canada, Bruce Patterson, is thanked for his diligence and assistance. He possesses patience beyond measure and tremendous editorial skill for which I am most grateful. This is quite aside from the 290 comments — many which were most amusing — and invaluable direction on not only questions related to heraldry and honours but the general content of this work. One is hard-pressed to think of anyone who suffered through reading such a large quantity of McCreery screed as Gauthier and Patterson.
Thanks are owed to General John de Chastelain for his continuing friendship and support. His provision of an insightful foreword for this edition is greatly appreciated. My parents, Paul and Sharon, along with Joyce Bryant and Peter Galloway, provided constant encouragement and professed interest in my work.
At St. John Ambulance headquarters in Ottawa, Dawn Roach and Patricia Kearney were helpful with questions about the Most Venerable Order. My good friend, Lieutenant-Commander Scott Nelson, was similarly helpful in editing one of the last iterations of the manuscript.
Alan Trammell of Pressed Metal Products and Stephen Hart of Rideau Ltée each provided important information in relation to questions about the manufacture of the Order of Canada and Order of Military Merit. Emmanuelle Sajous, Claire Boudreau, Cathy Bursey-Sabourin, Darcy DeMarsico, Denis Poirer, and Marie Glinski from the Chancellery of Honours at Rideau Hall were helpful in providing photo information and minute details about insignia and the internal operations of the Canadian honours system and Canadian Heraldic Authority. Darrel Kennedy, also from the Canadian Heraldic Authority, provided important comments on errors contained in the first edition. From the Department of National Defence, Juliane Martin and Corporal Jonathan Loeppky provided repeated assistance with images. Sam Kremer, outstanding collector of Champion Shot Medals, generously furnished this work with numerous scarce images.
Of my Ottawa friends, Glen Hodgins, Lieutenant-Colonel Dan MacKay, Kevin MacLeod, Mark Reid, Alana Blouin, Brigadier-General Christopher Thurrott, Brigadier-General Hilary Jaeger, and Mark O'Neill are each acknow-ledged for their support throughout this project and my other recent work, a history of the Rideau Club.
The cottage expedition crew — Sean and Lindsay Morency, Tara Bickis, Carl Gauthier, Melanie Nicholson, Jonathan Shanks, Jennifer Brown, Vasuda Sinha, Scott and Brenda Nelson — have all survived many trips in the tippy canoe K-225 and endured too many stories about medals and governors general. They are either very good at feigning interest in my work or are among the best friends one could hope for.
In Halifax I am grateful to the household staff at Government House, who are tolerant of my work habits and propensity for depositing empty teacups throughout the house. Lieutenant-Colonel Dale Warner and Captain(N) Craig Walkington remain keen to hear anecdotes about Canadian symbols over a variety of gin-based consumables, though this book may represent the limit of their tolerance. Matt Malone and Matt Follett provided many needed breaks from writing.
Robert Watt, Jacques Monet, Sheila-Marie Cook, Bishop Ralph Spence, Charles Robert, John Geiger, and John Fraser, all highly accomplished in their fields and public service, have regularly been free with advice and support throughout my writing and professional life. I feel fortunate to count them among my most trusted advisers.
As with other recent writing projects that I have embarked upon, my employer, Brigadier-General the Honourable J.J. Grant, lieutenant governor of Nova Scotia, and Mrs. Grant have offered much encouragement and friendship. My fellow vice-regal private secretaries from across Canada were also free with their encouragement and interest.
I remain indebted to Nancy Cameron. For more than thirty years my mother's secretary, "Aunt" Nancy gave me my first set of medals. From a numismatic hobby to a passion to study, it all started with a British War Medal and Victory Medal, my grade eight graduation gift from Nancy.
Christopher McCreery, MVO
Government House
Halifax
# Introduction to the Second Edition
A decade has passed since The Canadian Honours System was published by Dundurn Press, and since that time there have been a number of aadditions and changes to the Canadian honours system. I am grateful to the many readers who were kind enough to write and make suggestions, offer corrections, and encourage the production of a second edition. Over the period that has transpired since the release of the first edition, I have undertaken a significant amount of additional research into the history of the Order of the British Empire, the Order of Military Merit, the Royal Victorian Order, the Royal Canadian Mounted Police Long Service Medal, and the Canadian Forces' Decoration as the result of other publishing pro-jects. This research has yielded new and largely unknown information about the Canadian honours system and also necessitated a number of corrections. T. Robert Fowler's work Courage Rewarded: The Valour of Canadian Soldiers Under Fire 1900–2011, which examines the bestowal of gallantry decorations upon Canadians, has also significantly enhanced our understanding of this part of our honours system.
With the centennial of the First World War and seventy-fifth anniversary of the beginning of the Second World War, there is a heightened interest in the history of our honours system and how it has been used to recognize service men and women past and present. Although the honours from the two world wars of the past century are now historical artifacts, the surviving veterans of the Second World War give us a chance to learn about the stories that their insignia can tell. Similarly, the decade-long war in Afghanistan has also resulted in more than forty thousand Canadians being recognized with war service medals. This, too, has increased interest in the various forms of recognition accorded to our fellow citizens for service at home and abroad.
The most notable change to the Canadian honours system has been the physical manufacture of the Canadian Victoria Cross. Other important milestones have included the first conferral of the Star of Military Valour and Medal of Military Valour. It can now be said that Her Majesty's Canadian Armed Forces are served and recognized by a complete honours system.
A new addition to the Canadian honours system was the Sacrifice Medal, which was announced in September 2008, along with a number of new United Nations (U.N.) and North Atlantic Treaty Organization (NATO) Medals. The reorganization of the General Campaign Star and General Service Medal into awards with mission-specific ribbons — as opposed to a single medal with multiple bars — is another recent change to the honours system, along with the creation of rotation bars. Most recently, the addition of the Operational Service Medal and the Polar Medal has expanded the honours system further. Although considered mementos, the "memorial package" that is presented to the families of fallen members of the Canadian Armed Forces has been greatly enhanced to include elements of recognition previously given to those who died as a result of service in the First and Second World Wars. This, coupled with the Memorial Ribbon, makes the symbolic recognition accorded to the families of those who have made the ultimate sacrifice more extensive than at any point in Canada's military history.
This edition also includes chapters that examine the use of various imperial orders, decorations, and medals that were bestowed upon residents of Canada from the earliest times until 1967. Many readers of the first edition lamented the absence of a more robust overview of the pre-1967 honours bestowed on Canadians. I have used the term imperial as opposed to British to describe the pre-1967 honours conferred on Canadians because the honours were not unique to Canada. Although created by Royal Warrants of the Sovereign of the United Kingdom, nearly all of the pre-1967 honours examined in this work were awarded not only to Britons but also to New Zealanders, Australians, South Africans, Newfoundlanders, Indians, and many others.
As noted, this work provides an overview of the Canadian honours system; however, for an in-depth study of the policies and development of the modern Canadian honours system, readers should consult The Order of Canada: Its Origins, History, and Development, as well as a number of other works, also by the present author, related to specific Canadian orders, decorations, and medals:
* • On Her Majesty's Service: Royal Honours and Recognition in Canada
* • The Maple Leaf and the White Cross: A History of the Order of St. John and St. John Ambulance in Canada
* • The Canadian Forces' Decoration
* • The Order of Military Merit
* • Commemorative Medals of the Queen's Reign in Canada
* • Maintiens le Droit: Recognizing Service, a History of the RCMP Long Service Medal
In 2011 the author had the privilege of being commissioned by the Privy Council Office to undertake the most extensive review of the Canadian honours system to be completed since the reorganization of the Order of Canada and the establishment of the Order of Military Merit and the Decorations for Bravery. The ongoing adoption of many of the recommendations contained in the report has and will continue to result in a number of administrative changes to the way our honours system functions, along with the creation of a number of new honours, notably the Polar Medal and Sovereign's Medal for Volunteers.
An honours system is a living institution, one that changes with time. It expands and contracts depending on the needs of the society it is ultimately meant to serve on behalf of the Crown. Such a balance is not easily attained — even countries with ancient honours systems struggle with issues of accessibility, equality, insignia quality, and obtaining suitable recipients.
# PART I
# 1
ORDERS, DECORATIONS, AND MEDALS AROUND THE WORLD: UBIQUITOUS NATIONAL INSTITUTIONS
Like flags and coats of arms, almost every country possesses an honours system. At the centre of these systems there is usually a national honour. iBritain has the Order of the Garter and Order of the Thistle, France the Légion d'honneur, Japan the Order of the Rising Sun, Canada the Order of Canada, and the list goes on. Indeed, there are only a few countries that do not currently possess a national honour or honours system.[1] All other states and regimes have found a use for honours, and in many ways they are a necessity of nationhood.
The concept of honour is as old as humanity itself. Yet honours — that is, the official bestowal of recognition — are in comparison a relatively recent phenomenon. The concepts of honour and honours are directly linked, though there are important differences. Honour is the notion of adhering to what is right: high respect and good reputation through persistent good deeds. One can lead an honourable life and never receive official recognition in the form of an honour. Honours are official marks of recognition, whether they are conferred by proclamations, titles, grants of arms, or insignia. It is equally possible for a person to have received an honour and yet be dishonourable. These variables are defined by every society in accordance with the values that a particular civilization prizes.
Canada's honours system is derived primarily from two of its founding peoples, the French and British. However, not even these two countries can claim to have invented honours. Throughout modern history a plethora of systems and awards has been designed and used as essential apparatuses of both states and regimes. At times, systems have been displaced by both evolution and revolution, yet always a revised system, roughly patterned on its predecessor, has emerged. All modern honours systems have certain common elements, and the most central of these is the presentation of an insignia, being a badge or medal. This tradition can be traced back to circa 150 B.C. and the gold button given by Alexander, the ruler of Seleucid Syria, to a Jewish high priest for bravery demonstrated in battle. Honour has also been accorded through the bestowal of land and titles, and in ancient Greece a complex system of crowns was devised to reward public and military service. The practice of embossing medallic insignia on breastplates (phaleristics) was initiated by the ancient Romans and has in some ways carried forward to the present day. In India's Pudukkottai state there was an ancient tradition of the raja presenting gifts and honours to loyal subjects. This ritual served not only as a mechanism for recognition but was also a source of political power. As well, few are unfamiliar with the stories of King Arthur and his Knights of the Round Table. Moving through history, the ideals of chivalry and honours gradually replaced the act of granting land with that of bestowing knighthoods and insignia.
The purpose of honours is to reward service or valorous acts, and to accord recognition to those who, according to the state, are deemed to merit such. At various times honours have also been, and in some countries continue to be, a central tool of political patronage. All types of honours are intended to foster a feeling of loyalty and a personal connection to the state or regime.
Our modern national honours system can be divided into three main categories: orders, decorations, and medals. It is worth noting that there is some overlap between the last two categories, and some honours referred to as medals are actually decorations, and vice versa.
Orders
These are societies of honour that are instituted by the state, usually to recognize lifelong exemplary service of the highest calibre. Orders are usually divided into several different levels to allow for recognition of those who have rendered service at both the national/international level and at the local or regional level. In Canada we have several orders that are national in scope: the Order of Merit, the Order of Canada, the Order of Military Merit, the Order of Merit of the Police Forces, the Royal Victorian Order, and the Most Venerable Order of the Hospital of St. John of Jerusalem.
Orders can be "field specific" — the Order of Military Merit, for example, is restricted to members of the Canadian Armed Forces who have rendered meritorious service over an extended period of time. Similarly, the Royal Victorian Order is primarily bestowed upon those who have served the Queen or Royal Family in a distinguished manner.
Decorations
This section can be further divided into two subsections: Bravery/Valour/Gallantry and Meritorious Service. The term also applies to the Canadian Forces' Decoration and is colloquially used to describe all honours.
Bravery/Valour/Gallantry
Awarded for a specific act of bravery, valour, or gallantry, bravery awards are bestowed upon those who perform an exemplary act, such as life-saving, in a time of peace, while valour and gallantry decorations are for valour and devotion to duty in the presence of the enemy. Provisions have been made in the Canadian honours system to allow for valour decorations, such as the Victoria Cross (VC), to be awarded for gallant acts performed while in non-warlike situations, provided they involve a hostile armed force.
Meritorious Service
These are awarded for a specific act of meritorious service, not necessarily over an extended period of time. This category is notably employed to recognize short-term merit as part of a specific action or project, whether it is rendered over five minutes or five years.
Medals
There are four types of medals: service medals, commemorative medals, long service medals, and other awards. The term is often used in common parlance to describe all honours, be they the insignia of orders, decorations, or medals.
Service Medals
These are awarded for service in a particular mission or operation. While primarily limited to members of the armed forces, in the Canadian context police officers and civilian personnel from various government departments are also often included.
Commemorative Medals
Commemorative medals are awarded on the occasion of a special event such as a coronation, jubilee, or anniversary.
Long Service Medals
These are awarded for long service and honourable conduct over a set period of time. The Canadian Forces' Decoration (essentially a medal despite its name) is awarded for twelve years of honourable service, while the Royal Canadian Mounted Police (RCMP) Long Service Medal and Exemplary Service Medals are awarded for twenty years of honourable service.
Other Awards
Most specifically these are the Sacrifice Medal, awarded to those who have been physically or psychologically wounded as a result of service in an operational area; the Polar Medal, which is awarded for specific service in Canada's North; and the Queen's Medal for Champion Shot, which is awarded for winning an annual marksmanship competition. In the provinces this category includes Volunteer and Citizenship Medals.
Waterloo Medal obverse.
Waterloo Medal reverse.
These three main types of honours emulate directly the British system of honours, which, although ancient in its origins, only began to develop in this form in the mid-nineteenth century. Since the fourteenth century, England had used its Order of the Garter and other knighthoods to reward loyal nobles. In the nineteenth century the Order of the Bath was expanded, while others such as the Order of St. Michael and St. George, the Order of the Star of India, and the Order of the Indian Empire were created. These were not restricted to nobles, and their membership came to include non-titled members of the military and civil service.
The Battle of Waterloo yielded the first standard-issue medal intended for wear. Created in 1816 at the direction of the Duke of Wellington, the circular medal measured 36 mm in diameter with a ring suspender hung from a ribbon and served, in many ways, as the basis for the design of future war medals. There was the additional aspect that the medal was impressed with the recipient's name and issued to officers and men alike.
In 1830 and 1831 respectively, the British Army and Royal Navy each instituted long service medals. The Army Long Service and Good Conduct Medal was awarded for twenty-one years of service in the ranks. The Royal Navy Long Service and Good Conduct Medal was originally awarded for twenty-one years of service as a rating in the Royal Navy. These were the first long service awards, and both continue to be awarded to members of the British Army and Royal Navy.
At various times members of the Canadian military were awarded a diverse number of long service medals and decorations. These were dependent not only on the length of service rendered but also the branch of service, rank — non-commissioned officer or commissioned officer — and whether one was in the regular force or reserve. In all there were more than a dozen different long service awards consolidated into the Canadian Forces' Decoration in 1951. With the Canadian Forces' Decoration, the distinction between service in the regular and reserve forces was abandoned, as was having separate awards for officers and non-commissioned officers.
In 1847 the Military General Service Medal was established, and in the following year the Naval General Service Medal was created to recognize service rendered in specific wars and actions between 1793 and 1840. These early awards have served as the basis for modern service medals throughout the Commonwealth: their standard design, bearing the Sovereign on the obverse and an allegory on the reverse, is a tradition that continues, most recently with the institution of the Polar Medal and the Operational Service Medal.
The first standardized bravery decoration was created in 1854 with the institution of the Distinguished Conduct Medal, followed two years later by the Victoria Cross. The VC was quite a novel development in that it was open to all ranks from private to field marshal. During the First World War, a broad range of branch specific awards for the navy, army, and air force were established.
Canada is not alone in taking much of its honours structure from Britain. In an evolutionary sense it was both logical and practical. Countries from Australia to Zimbabwe all find the general basis of their honours system in the British tradition, at least in terms of structure. Australia, Fiji, and certain other Commonwealth countries also derive the structure of their national orders from the Canadian experience.
Although influenced by the British tradition, the Canadian honours system is not simply the imperial honours system covered with maple leaves and fleurs-de-lys. There are significant differences between the Canadian and British honours systems, and between the Canadian honours system and those used in many other countries. The Canadian system is non-partisan — that is, members of the governing party are neither directly nor indirectly involved in the selection of recipients of honours. In many countries such decisions are made by the head of government (e.g., the prime minister), and thus honours invariably become a political patronage tool. In many countries it is only members of the political and social elite who are found with neck badges and medals, but not so in Canada. Canada is also one of the few countries that does not engage in automatic awards.
The focus of this study is honours that are officially sanctioned and created by the Crown at the federal and provincial levels in Canada. Official honours in Canada emanate from the Queen, who is the country's head of state (worldwide, most national honours are created by an instrument approved by the head of state). In the case of the provinces, various honours have come into being after the granting of Royal Assent by the lieutenant governor, as personal representative of the Queen, to legislation passed by the assembly or through an order-in-council that the lieutenant governor, on behalf of the Crown, has assented to. These provincial honours are therefore deemed to emanate from the Crown in right of that particular province. Nevertheless, only when these provincial creations are incorporated into the national Order of Precedence for the wearing of orders, decorations, or medals can it be said that they have been officially recognized by the Queen in Right of Canada and can therefore be worn along with other official honours from the Crown.
Similarly, other honours such as U.N. and NATO service medals come from recognized head of state level organizations that the Crown in Right of Canada has agreed to recognize, along with their associated rules and regulations, and incorporate into the national Order of Precedence so they can be worn with official Canadian honours.
These incorporations in the official order of wear, however, do not make these honours, either provincial or organizational, formally part of the Canadian honours system, nor do they imply that they are awarded by or on behalf of the Queen. Only those honours that are created by and awarded on behalf of the Queen in Right of Canada are part of the Canadian honours system, while the Order of Precedence allows for the official wearing of a wider array of honours that are "recognized" by the Crown of Canada.
Unofficial honours (see chapter 39), that is, those honours not recognized by the Government of Canada, including provincial (and even municipal) honours that have not been sanctioned for wear by the Government of Canada, are only peripherally referenced herein.
# 2
HONOURS IN CANADA FROM THE EARLIEST TIMES
Honours in Canada to 1759
The modern Canadian honours system can find some of its origins in those of two of Canada's founding peoples: the French and British honours systems. Canada's Native peoples did not have a formal honours system in the sense of an institution. Rather, the concept of honour was more prized than any particular insignia or possession. Service to the community and bravery were recognized in Native communities not by gold or silver medals but through respect accorded to the individual often through the adoption of a chieftainship. When Europeans arrived, this system was complemented through the awarding of Chief's Medals, which French and British officials both used to secure the loyalty of a particular Native group. This process was not unidirectional; Native peoples quickly learned how to extract better terms and conditions for themselves when accepting such medals from Europeans.
The Honours of Royal France in Canada
Prior to the French Revolution, the Kingdom of France possessed a comprehensive honours system, primarily graded by the hierarchy of French society at the time. The king of France would elevate French subjects to the nobility as dukes, marquises, counts, viscounts, and barons. On one occasion a Canadian, Charles Le Moyne, Seigneur of Longueuil, was made Baron de Longueuil in recognition of his peace negotiations with the Iroquois on behalf of the French Crown. Le Moyne would later go on to serve as governor of Montreal. This now-ancient French title still exists, though the family no longer lives in Canada.
In New France honours were awarded by the governor on behalf of the king. Prior to an honour being bestowed, the governor of New France had to receive permission from the French sovereign. The advice of the governor general was usually followed, though there were occasions when the king of France bestowed an honour without the prior knowledge of the governor. This was, after all, his prerogative as king.
Commander of the Order of St. Louis.
Looking for a way to motivate and reward all the officers of his armies for outstanding and long services, Louis XIV established the Ordre royal et militaire de Saint-Louis in April 1693. For the subjects of New France this was the most familiar honour, and it became an immediate success. The order gained the favour of all and was seen as a very prestigious honour, since the deserving officer would wear the same insignia as worn by a marshal of France or, even better, the king himself! It was the first military reward in modern history open to officers in the army or navy regardless of social status. However, it was awarded only to Roman Catholic officers. The order consisted of three classes: Knight (with no limit on overall membership), Commander (with a maximum of twenty-four living members), and Grand Cross (with a maximum of eight living members).[1] After the French Revolution, the order was abolished by the Convention in 1792, though it continued to be used by the government in exile and was revived following the first Restoration and finally discontinued in 1830 by King Louis-Philippe. Nearly three hundred appointments to the order were made to people associated with Canada, either French-born officers serving in New France, Canadian-born officers serving in New France, or Canadian-born officers serving in France during or after the fall of New France.
Comte de Frontenac, Grand Cross of the Order of St. Louis, governor general of New France.
The first officer to be appointed to the order was the governor of Montreal, Louis-Hector de Callière, in 1694. Louis de Buade, comte de Frontenac and governor of New France, was the second, in 1697. The best-known Canadian subject to receive the Order of St. Louis was François Coulon de Villers. Born in Verchéres, New France, in 1712, Coulon de Villers went on to serve with great distinction in the French colonial army. In the battle for Fort Necessity on July 4, 1757, Coulon de Villers became the only man to ever defeat George Washington in battle. The last governor general of New France, Pierre de Rigaud, Marquis de Vaudreuil-Cavagnial, was the only Canadian to receive the Grand Cross.
With the fall of New France in 1759, one would expect the flow of honours from France to come to a halt. However, this was not the case, since up until the French Revolution the Crown continued to bestow the Order of St. Louis upon residents of the old colony in recognition of military services rendered prior to 1759. These awards were usually given to veterans whose families had petitioned for belated recognition.
# 3
IMPERIAL HONOURS IN CANADA FROM CONFEDERATION TO THE CENTENNIAL
To an outside observer it might be strange to include what some would consider a foreign honours system in a book that examines Canadian honours. This would be true if not for the fact that until recently most orders, decorations, and medals awarded to Canadians were borrowed from the British. While Canadians administered these honours and Canadian governments had control over who received them, they were for the most part common throughout the British Empire/Commonwealth. Indeed, while Canada ceased using British honours in 1967 with the creation of the Order of Canada, some Commonwealth countries such as Australia and New Zealand have only followed the Canadian lead within the past forty years. There remain a number of the Queen's smaller realms such as Papua New Guinea, the Solomon Islands, Belize, Tuvalu, Grenada, and St. Lucia that continue to use imperial honours or what they now call "the Queen's honours." In some cases these are used alone or in tandem with local national honours.
The evolution of honours systems in the Commonwealth is a fascinating subject, one that relates closely to the development of Canada's modern honours system. Countries such as Australia, Fiji, and the Solomon Islands have adopted national orders in part based on the Order of Canada. The Canadian experience with developing a national honours system has been closely followed by many countries keen on creating an indigenous honours system while continuing to respect that which came before.
Given the large number of Canadians who received imperial honours during the past century, many of whom are still alive, it would be recklessly revisionist to pretend that such honours were foreign.
The term imperial honours is used throughout this book, not out of some antediluvian attachment to the words but because the British honours system in Canada and the Commonwealth was until relatively recently an imperial honours system, one tacitly administered by the various government departments responsible for honours in the United Kingdom, but one in which there was a high degree of local control over the type of awards and who was honoured. It is misleading to consider them to be only "British" when they were, in fact, a system shared, shaped, and used by the whole of the Commonwealth at one time or another.
The administration of each order, decoration, and medal covered in this section is discussed in the following six chapters. Canadian officials had a significant amount of control over who would be nominated for or receive a specific award. From an accountant's point of view it was an ideal situation, since the British defrayed most of the costs for insignia and certificates while Canadian officials merely had to cover the postage charge for the shipping of insignia.
For senior honours, administration was left to the prime minister, who was allotted a certain number of knighthoods for which he could recommend eminent Canadians. While the Sovereign was — and remains — the "fountain of all honours," the prime minister once had a great degree of personal control over who received honours, whereas today he or she has no formal influence over who is recognized by the Crown in Canada. Gallantry awards were non-political and thus there was no interference by political officials, though they were occasionally involved in the broader process of sitting on committees and sending along recommendations for bravery awards to the Department of the Secretary of State, which was, until 1972, charged with the administration of honours in Canada.
As we will see, aside from a few very high-profile cases, the Canadian government once had an immense degree of control over the flow of honours. Nevertheless, the perception remained that imperial honours were tightly controlled by the British government. This may have had some truth to it in connection to peerages and certain knighthoods conferred upon Canadians, but in terms of those honours Canadians became most familiar with, such as the Order of the British Empire, war service medals, long service medals, and the Imperial Service Medal, Canada had near-complete autonomy.
Canada continues to differentiate between Commonwealth and foreign honours. Because members of the Commonwealth recognize the Queen as head of the Commonwealth, honours emanating from a Commonwealth country are given precedence over those from non-Commonwealth countries. This is closely connected with the legal reality that Commonwealth countries are not considered to be foreign, because of the fact that we share a common link through our history and the Sovereign.
The section on imperial honours does not cover medals such as the Naval General Service Medal, 1915–62, and the Military General Service Medal, 1918–62, despite the fact that a small number of Canadians received these while on exchange with the British Armed Forces. Similarly, the Falkland Islands War Medal, the Campaign Service Medal, and most recently the British Operational Service Medal (for Sierra Leone) have been awarded to Canadians; however, their numbers remain so low that inclusion is not warranted here. Today the last three medals are considered "Commonwealth" and not Canadian.
The Imperial Honours System in Canada
In Canada the Crown has always been the "fountain of all honours." As such all officially recognized national and provincial honours must be created by the Sovereign and either awarded in the name of the Queen or sanctioned by the Crown. The bestowal of honours in Canada was articulated through the honours list, originally published in the London Gazette and later the Canada Gazette. During peacetime, the list was published twice yearly, once on the Sovereign's birthday and once around Christmas or New Year's Day. For Canadian civilians these lists were compiled by the governor general and prime minister — though the level of their co-operation varied — and then submitted to the Sovereign for approval. For members of the Canadian military, honours lists were drawn up by senior military officers and the minister of national defence, then approved or augmented by the governor general. Until 1918 the British government had the power to nominate Canadians for honours, though this prerogative effectively came to an end with the Nickle Resolution.
Prior to Confederation the British government did not feel obligated to consult the colonial governments of British North America before conferring an honour upon a resident of what would later become the Dominion of Canada. The system was quite simple: the governor or governor general would suggest to the colonial secretary that a particular person be recognized; if the colonial secretary approved the nomination, it would be put forward for the approval of the British prime minister and ultimately the Sovereign. Recommendations did not have to originate with the governor or governor general; in theory they could originate from any member of the general public, though in practice it was unlikely that the colonial secretary would act upon the recommendation of, say, a Mr. Shanks of Bytown, Canada West. Similarly, the colonial secretary could suggest that a particular person in a colony be recognized, and of course the governor of the colony in question, as the local authority, would normally be consulted.
Residents of Canada were eligible for most British honours, save those that did not apply to Canada, such as the Order of the Star of India or the Order of the Indian Empire. In the pre-Confederation period relatively few honours were bestowed upon Canadians. Those awarded were usually knighthoods conferred on judges and senior politicians. There were no junior civil service honours or long service medals for the local constable.
An Emerging Policy, 1867–1917
Cartoon of Sir Wilfrid Laurier, PC, GCMG, shortly after his dubbing in 1897.
After Confederation a convention emerged whereby the prime minister of Canada submitted his honours lists to the governor general, who vetted them and submitted them to the colonial secretary and then onto the Sovereign. The governor general — who was then a British official — also nominated Canadians for honours, usually without the knowledge of the Canadian prime minister. Awards for members of the military were submitted by the general officer in command of the Canadian Militia to the governor general for transmission to London, though there were some instances when the prime minister nominated senior Canadian officers for honours.
This system of informal consultation functioned fairly well until 1901 when controversy arose over the knighting of Thomas Shaughnessy, president of Canadian Pacific Railway (CPR). Governor General Lord Minto suggested to Prime Minister Sir Wilfrid Laurier that Shaughnessy be knighted on account of his services during the 1901 Royal Tour of the Duke and Duchess of Cornwall and York (the future King George V and Queen Mary). Laurier opposed the idea on the grounds that Shaughnessy was unpopular with Canadians and certainly no friend of the prime minister. Minto, however, disregarded Laurier's advice and advanced the nomination.
Laurier was furious when Shaughnessy was knighted, and it did not help that he learned of the appointment from a newspaper and not the governor general. By 1902, Laurier had drafted an official policy on honours in Canada. It set out that all honours, save the Royal Victorian Order, had to be approved by the prime minister before any list could be sent from the governor general to the King. The governor general and the British government took their time in replying to Laurier's policy, and while they agreed that the prime minister should be involved in reviewing the honours lists and submitting names, they maintained that the governor general would retain the right to nominate Canadians.
Discontent and Discord: The Nickle Debates, 1917–19
The First World War brought much social and constitutional change to Canada, which included the nation's policy toward honours. Prior to the war there had been only muted opposition to titular honours such as peerages and knighthoods. Honours were viewed as necessary, even if they were used as tools of patronage from time to time. This prevailing attitude changed during the Great War as a result of several high-profile controversies. In 1914 a private member's bill was introduced in the House of Commons that sought to abolish peerages and knighthoods in Canada, but not other honours. This proposal was poorly received and promptly defeated. By 1917, however, the mood had changed.
In particular, two very public scandals over honours induced Parliament to examine the issue. The first involved the 1915 appointment of the Canadian minister of militia and defence, Sam Hughes, as a Knight Commander of the Order of the Bath. Hughes had been pilloried in the press because of his overall administration of the department of militia and defence and the Canadian Expeditionary Force. His bombastic approach, cronyism, and procurement of the Ross Rifle were widely reported. The more serious outrage occurred in 1917 when Sir Hugh Graham, owner of the Montreal Star and a staunch imperialist, was elevated to the peerage as Lord Atholstan — against the advice of the Canadian prime minister and governor general. Graham's peerage was so controversial in part because of his highly unpopular right-wing views and because of an increasing suspicion among Canadians of peerages and knighthoods. Although an unpopular figure, Graham had built up a major newspaper and founded a number of other publications.
This was the only time in Canadian history that the British government ignored advice from both a governor general and a prime minister. The disregard of Ottawa's wishes was related to the fact that British Prime Minister David Lloyd George was selling peerages and knighthoods to raise funds for his party. While the use of honours to reward political contributions was hardly new, under Lloyd George such transactions had become particularly blatant, as is noted below.
In addition to these events, there was an underlying naïveté about honours in Canada. Peerages and knighthoods were thought to be the same thing — both hereditary — and there was similar confusion regarding the imperial orders of chivalry. For instance, when the creation of the Order of the British Empire was announced in 1917, Canadian newspapers announced that three hundred Canadians were going to be knighted with the new order. This was certainly not the case.
Sir Robert Borden, PC, GCMG, KC.
In March 1917, following Graham's elevation to the peerage as Lord Atholstan, Sir Robert Borden drafted a new government policy setting out that all honours must be approved by the Canadian prime minister and that no further hereditary honours (peerages or baronetcies) were to be conferred on Canadians. Only a week after this policy was drafted, William Folger Nickle, the Conservative-Unionist Member of Parliament (MP) for Kingston, introduced a resolution in the House of Commons requesting that the King cease awarding peerages to Canadians. Nickle had no trouble with knighthoods or other honours, only those that had a hereditary quality. Nickle's resolution was, in fact, very similar to Borden's new policy. After a lengthy debate, the House of Commons adopted a resolution placing power over recommendations for all honours in the hands of the Canadian prime minister, while at the same time asking the King to cease awarding hereditary titles to Canadians. This is what came to be known as the Nickle Resolution, even though Nickle himself voted against the version eventually adopted.
Although the Nickle Resolution was adopted, the debate was far from over. While military honours continued to be conferred, Borden did not send forward any further recommendations for honours. He thought the issue was still too contentious to test the new protocol.
Throughout late 1918 and most of 1919, the British press was littered with reports about people purchasing honours. Although this was a problem confined to Britain, many people in Canada assumed that the same practice was followed on this side of the Atlantic as well. Fearing that an avalanche of knighthoods was to accompany the newly created Order of the British Empire, Nickle introduced another motion in April 1919. This one called for the King to "hereafter be graciously pleased to refrain from conferring any titles upon your subjects domiciled or living in Canada."
William Folger Nickle, KC, MP.
Nickle was now pursuing a prohibition on all titular honours, a departure from his original opposition to only hereditary honours. Following another lengthy debate that in many ways mirrored the one in 1918, the House of Commons voted to create a Special Committee on Honours and Titles, which held several meetings and eventually submitted a report to Parliament that called for the King to cease conferring all honours and titular distinctions, save military ranks and vocational and professional titles, upon residents of Canada. It also recommended that action be taken to extinguish the heritable quality of peerages and baronetcies held by Canadians, something that would be impossible to do. The committee approved of the continuance of naval and military decorations for valour and gallantry. The final part of the report affirmed the committee's desire to see that no resident of Canada be permitted to accept a title of honour or titular distinction from a foreign (non-British) government. The Commons passed a motion of concurrence with the report and it was adopted.
There has, invariably, been confusion about the Nickle Resolution and the Report of the Special Committee on Honours and Titles. Neither was a statute, and neither had any standing as anything more than a recommendation or guideline, as Prime Minister R.B. Bennett demonstrated in 1933. The Nickle Resolution served as a policy document on how a prime minister could submit honours lists, and while it requested that no further hereditary honours be bestowed, it did not prevent Canadians from accepting other honours, whether a knighthood or Officer of the Order of the British Empire (OBE).
Prohibition, 1919–1932
The prohibition on Canadians accepting imperial honours commenced in 1918 and lasted until 1933. Neither Prime Minister Arthur Meighen nor his successor William Lyon Mackenzie King submitted honours lists, though it was well within their power to do so. Both leaders hesitated on account of the issue's contentiousness. The prohibition was not complete, however, as Canadians living in other parts of the British Empire were still eligible to be awarded honours. In 1925 Canadian-born Emma Albani, the world-renowned soprano and the first Canadian to attain international stardom as a result of artistic abilities, was honoured with a damehood. Albani was made a Dame of the Order of the British Empire in 1925, the first Canadian woman to be so honoured, but since she resided in the United Kingdom, the Canadian government did not become involved in protesting her appointment.
Another prominent example was that of Dr. Sir George Washington Badgerow, a famous ear, nose, and throat doctor. Badgerow was born and trained in Canada, though he made his home in Britain. In 1926 the British government requested permission from the Canadian government to allow Badgerow to be knighted. Two years later the Canadian government responded that the award could go forward because Badgerow, though born in Canada, was a resident of Britain and was being rewarded for services he performed in Britain. Thus the prohibition was incidental, and Bennett would prove that there was, in fact, no legislative prohibition at all, but rather a series of prime ministers who had no interest in honours lists.
Bennett's Honours Lists, 1932–1935
Prime Minister R.B. Bennett broke the moratorium on honours that had existed in Canada between 1919 and 1933. In fact, Bennett adhered perfectly to the Nickle Resolution and had eighteen Canadians awarded knighthoods and 189 appointed to the various non-titular levels of the imperial orders of chivalry. Bennett solicited nominations from the various lieutenant governors and other officials and then personally selected each candidate.
Richard Bedford Bennett, PC, ED, KC.
Unlike previous lists, Bennett's were largely non-partisan and well distributed among the provinces and between both sexes — quite a novelty for the period.
Among others, Bennett's lists recognized Sir Frederick Banting, the co-discoverer of insulin; Sir Ernest Macmillan, the noted composer and conductor; Sir Thomas Chapais, the esteemed historian; and Sir Arthur Doughty, the dominion archivist. At the non-titular level, Lucy Maud Montgomery was made a Commander of the Order of the British Empire (CBE) and Lester Pearson was made an Officer of the Order of the British Empire (OBE). For the first time women represented nearly half of those being recognized with honours. Public reaction to these awards was ambivalent.
Bennett himself was offered a Knight Grand Cross of the Order of St. Michael and St. George (GCMG), though he declined it and offered it to William Lyon Mackenzie King, who also declined it. Bennett was the last Canadian prime minister to make use of all aspects of the imperial honours system, with the final knighthood made on the recommendation of a Canadian ministry being a Knight Commander of the Order of St. Michael and St. George (KCMG), going to one of Canada's first diplomats, the Honourable Herbert Meredith Marler, Canada's minister plenipotentiary to Japan.
The Second World War
When Mackenzie King was returned as prime minister in 1935, the brief revival of imperial honours came to an abrupt end. Thus, Canada entered the Second World War with no policy on honours, other than to allow for the bestowal of gallantry and valour decorations. Even this policy was unclear as to who would approve the awards, from where nominations would emanate, and how many were to be allotted for Canadian service personnel. To deal with these problems a special interdepartmental committee, which would later become the Awards Coordination Committee (ACC), was founded in early 1940. This committee drafted formal honours policies, the first of which set out that Canadians could receive gallantry decorations "in operations against the enemy."[1] No provisions were made for appointing Canadians to the various imperial orders of chivalry.
William Lyon Mackenzie King, PC, OM, CMG.
Throughout the Second World War, numerous proposals were devised to create a Canadian order, but none came to fruition, and thus Canada continued to work within the broader imperial honours system. In 1942, Parliament again tackled the issue of honours and awards. In July of that year, the Awards Coordination Committee met and ultimately decided that, in addition to being allowed to receive gallantry and valour decorations, Canadians should be eligible for the non-titular levels of the imperial orders of chivalry (those not conferring knighthood). It also expressed the opinion that the Canadian government should establish a Canadian order of merit. Essentially, the committee adhered to the Nickle Resolution — which allowed for the award of honours only on the advice of the Canadian government — while simultaneously discarding the proposals of the 1919 Special Committee on Honours and Titles, which sought to cease the naming of Canadians to all imperial orders of chivalry.
It was not until 1967 with the founding of the Order of Canada that the Canadian government acted upon the 1942 proposal to create a Canadian honour. In terms of the other proposals, Canadian service personnel and civilians were made eligible for the non-titular levels of the Order of the Bath, the Order of St. Michael and St. George, the Order of the British Empire, and the Imperial Service Order. This came to an end with the conclusion of the war, and Dominion Day 1946 saw the last civil honours list until the establishment of the Order of Canada.
The Korean War
With the beginning of the Korean War, Cabinet was once again faced with making decisions about honours in Canada. It elected merely to institute a policy similar to that used during the Second World War, which allowed for members of the Royal Canadian Navy, Canadian Army, and Royal Canadian Air Force to accept imperial gallantry decorations and the non-titular levels of the imperial orders of chivalry. However, in contrast to the Second World War, civilians who made an important contribution to the Korean War effort were not permitted to receive any honours.
The 1956 Policy: An Incomplete Solution, 1956–1967
The end of the Korean War saw the flow of imperial honours, aside from long service awards, cease once again. In 1956, however, Cabinet passed a directive that instated a more liberal policy toward honours and awards. Cabinet Directive 30 allowed Canadian civilians to receive the George Cross, George Medal, and Queen's Commendation for Brave Conduct for "acts of bravery performed at the risk of death or serious injury." Thus, for the most part, the Canadian honours system consisted of bravery decorations. Allowances were also made to permit appointments for gallantry to the various levels of the Order of the British Empire, and a handful of appointments to the order were made to both civilians and members of the Canadian Armed Forces for bravery. Members of the military were still eligible for the various imperial gallantry decorations, though there were no such allowances for other recognition (see chapter 5). The 1956 policy also allowed for Canadians to accept foreign honours under certain circumstances. This was the first time since the Second World War that Canadian civilians were permitted to accept foreign honours.
The final Canadian to be awarded an imperial honour on the advice of the Canadian government came in 1968 when Captain (Nursing Sister) Josephine A. Cashin of the Royal Canadian Army Medical Corps was made a Member of the Order of the British Empire (MBE) with gallantry emblem. This was in recognition of her bravery in assisting with the rescue effort following the crash of a Czechoslovakian State Airline plane near Gander International Airport in September 1967.
Establishing the Canadian Honours System, 1967
Letters patent constituting the Order of Military Merit, 1972.
The modern Canadian honours system finds its basis, initially, in the Order of Canada, established by letters patent signed by Queen Elizabeth II on March 17, 1967. At first the order consisted of three separate parts: Companion of the Order of Canada, the Medal of Courage of the Order of Canada, and the Medal of Service of the Order of Canada. These three awards served as the cornerstone upon which the post-1967 Canadian honours system has developed. The development of the Order of Canada has embodied much more than the creation of a single institution; it is rather a project that expanded to include the various Canadian Decorations for Bravery, the Order of Military Merit, and an entire family of honours. To this has been added the provincial orders and a select number of provincial medals.
Honours in the Dominion of Newfoundland
Newfoundland, itself a periodic autonomous dominion prior to joining Canada in 1949, had a rather different honours experience. Indeed, imperial honours were bestowed upon Newfoundlanders right up until the eve of Confederation. There was never a prohibition on residents of this dominion from accepting any honours, and liberal use was made of the system, especially between 1930 and 1949. The governor and premier played a direct role in drawing up nomination lists that were submitted to the Colonial and later the Dominions Office.
During the Commission of Government period (1934–1949), when there was no premier, the roles of the governor and chairman of the commission were paramount in the process. As with the appointments made to Canadians during the Bennett premiership and throughout the Second World War, the Dominions Office simply provided each dominion with a maximum allotment of honours, and they did not become involved in the actual nomination process. Much to the annoyance of the Canadian government, the flow of knighthoods to Newfoundlanders continued right up to the moment of Confederation.
On a per capita basis, Newfoundland was one of the most honoured parts of the British Empire, its residents having received one peerage, one Grand Cross of the Order of St. Michael and St. George (GCMG), thirteen Knight Commanders of the Order of St. Michael and St. George (KCMGs), fourteen Companions of the Order of St. Michael and St. George (CMGs), nine Knight Commanders of the Order of the British Empire (KBEs), one Dame Commander of the Order of the British Empire (DBE), three military Commanders of the Order of the British Empire (CBEs), twenty-six civil CBEs, nine military Officers of the Order of the British Empire (OBEs), forty-eight civil OBEs, ninety-eight Members of the Order of the British Empire (MBEs), and one Imperial Service Order (ISO). One Newfoundlander serving in the Royal Navy was a Knight Commander of the Order of the Bath (KCB) and later a Grand Cross of the Order of the Bath (GCB), while his brother was a Companion of the Order of the Bath (CB) — all in the military division.[2]
The Creation of Canadian Honours Today
Proposals to create new honours or medals generally come from the Chancellery of Honours or the Department of National Defence, though they occasionally emanate from individuals, organizations, members of the public, or commissioned reports.
The Chancellery researches to ensure that the proposed honour or medal is needed and that it conforms to Canadian honours policy. Once it deems that the honour is needed, a proposal is presented to the Honours Policy Committee, a government body chaired by the clerk of the Privy Council and made up of senior public servants from various federal departments. If the Honours Policy Committee agrees that the honour is needed, the Chancellery is then directed to propose regulations and a draft order-in-council. Design proposals are also drawn up by the Chancellery, usually in collaboration with the Canadian Heraldic Authority.
The Honours Policy Committee then reviews the proposal again and, if approved, it is passed on to the prime minister for consideration. If the prime minister approves the new award, the Queen's informal agreement is sought. From this point an order-in-council is processed through the Privy Council Office. On the advice of the prime minister of Canada, through the governor general's office, the letters patent and design paintings are sent to Buckingham Palace for approval by Her Majesty the Queen. It is only when the Queen signs the letters patent that the honour is considered to be officially created.
Once the paintings and letters patent are signed by the Queen, a press release is sent out by the governor general and the regulations for the new award are published in the Canada Gazette. Amendments to the criteria for Canadian orders, decorations, and medals do not have to be approved by the Queen but can be made by the governor general.
# 4
PEERAGES, BARONETCIES, AND THE IMPERIAL ORDERS OF CHIVALRY
While only infrequently summoned to the peerage of the United Kingdom — thus becoming a member of the House of Lords — a number of prominent Canadians have been elevated to the upper chamber of the Parliament at Westminster in connection with their services to the British Empire, Britain, or Canada. A peerage not only constitutes a hereditary title and a seat in the House of Lords but it was and remains considered an honour of the Crown. A baronetcy, a sort of hereditary knighthood that does not entitle the holder to sit in the House of Lords, was also employed as an honour of the Crown, though it was bestowed on only a few Canadians.
Residents of Canada were more frequently appointed to the various imperial orders of chivalry. Indeed, the diverse array of orders served as the principal means of bestowing official recognition upon Canadians for meritorious achievement and lifetime achievement in the military and civil fields until 1967. This chapter provides an overview of the type of service each order was used to reward, the number of Canadian residents appointed to each order, and various details related to the insignia recipients were presented with.
Canadian citizens continue to be appointed to a number of these orders, though solely for services to the United Kingdom or a number of the Queen's other realms. As there are a large number of Canadian citizens who hold British citizenship and there are continuing close relations between the two countries, it is not uncommon for Canadians to be rewarded with appointment to an imperial order in recognition of services rendered to the United Kingdom.
Honours such as the Imperial Service Order are no longer awarded to British citizens, though through a curious arrangement citizens of a number of Commonwealth countries continue to be appointed to this order on the advice of their national governments.
The Order of Merit, Royal Victorian Chain, Royal Victorian Order, and Order of St. John are all examined in Part II. They are elements of the Canadian honours system that are shared with the United Kingdom and the Queen's other realms, whereas the peerage, baronetcy, and orders examined in this chapter clearly fall into the category of imperial honours that Canadians have not been appointed to on the advice of the Government of Canada (Queen in Right of Canada) since 1968.
A number of imperial orders of chivalry are not detailed in this chapter. These include the Most Noble Order of the Garter, the Most Ancient and Most Noble Order of the Thistle, the Most Illustrious Order of St. Patrick, the Most Exalted Order of the Star of India, the Most Eminent Order of the Indian Empire, the Royal Guelphic Order, the Imperial Order of the Crown of India, and the Order of Burma. While many of Canada's governors general and their spouses were members of some of these orders, few, if any, were bestowed upon residents of Canada. Only one person closely associated with Canada, Lord Pirrie (born in Quebec City), was appointed to the Order of St. Patrick. Former Governor General Vincent Massey would have been made a Knight of the Garter if the Queen had not been advised against doing so by Prime Minister Louis St. Laurent and his successor John Diefenbaker. The Royal Guelphic Order has a connection to Canada through a number of important figures, notably Sir Francis Bond Head, lieutenant governor of Upper Canada, and Sir John Franklin, the intrepid Arctic explorer who perished in what is modern-day Nunavut. Unless otherwise stated, all of the numbers listed in this chapter are for honours bestowed upon residents of Canada for services to Canada. Where possible the number of honours bestowed on Newfoundlanders has also been listed.
The Peerage in the United Kingdom
The tradition of bestowing titles upon loyal subjects of the Crown in Canada was not unique to the British. Under the French regime, titles were bestowed upon particularly distinguished subjects (see chapter 3).
A map of this complex system begins with the peerage in the United Kingdom, which is made up of different levels of peers:
> I Duke/Duchess
>
> II Marquess/Marchioness
>
> III Earl/Countess
>
> IV Viscount/Viscountess
>
> V Baron/Baroness
>
> VI Life Peers (Baron/Baroness)
Prior to 1999, almost all members of the peerage were entitled to sit in the House of Lords, the upper chamber of the British Parliament — much like the Senate is the upper chamber of Canada's Parliament. Until 1957 almost all peerages were hereditary; thus the title passed to the eldest male heir of a peer upon his death. In some exceptional cases, a peerage title could pass through a female child, as happened with Lord Strathcona, who had two daughters but no son.
The peerage is unusual in that being summoned as a peer is both an honour — in that a title is bestowed — and that, until recently, one generally became a member of the upper chamber of the British Parliament. The passage of the House of Lords Act, 1999, removed all but a handful of hereditary peers from the House of Lords, making it more like the Canadian Senate. Since 1957, almost all appointments to the peerage have been of life peers, meaning the title (always a baron) is held only by the recipient and is not inherited after his or her death.
Over the span of Canadian history, twelve peerages and one other title have been bestowed upon Canadians. Among these, five are considered to be Canadian peerages — that is, awarded after consultation with the Canadian government. The first of these went to Sir George Stephen, who became a baron on June 26, 1891. Stephen, like two other Canadian peers, was president of Canadian Pacific Railway, and it is not unlikely that he owed his appointment to the "help" he gave the ruling Conservatives in the 1891 election campaign. This connection between the CPR and the awarding of titles strengthened opposition to hereditary honours among some in the Liberal Party of the period, who connected the procedure to the operations of patronage and party financing in the Canadian political system.
The second Canadian (and sole Canadian woman) appointed to the peerage, Lady Macdonald, became Baroness Macdonald of Earnscliffe on August 14, 1891. She was not entitled to sit in the House of Lords, and it seems safe to conclude that she was honoured mainly in memory of her late husband, Prime Minister Sir John A. Macdonald, who was to be the initial recipient of the honour had he not died while still an MP.
The next appointment to the peerage, Sir Donald Smith, was another former president of the CPR. He was appointed in his capacity as Canadian high commissioner to London, becoming 1st Baron Strathcona and Mount Royal in 1900. Smith continued as Canadian high commissioner until his death in 1914. Sir Thomas Shaughnessy, the principal catalyst for the 1902 Privy Council Report on honours and titles — and like Strathcona and Stephen also a president of the CPR — was elevated to the peerage in 1916.
The last Canadian to be elevated to the peerage after consultation with the Canadian government was Sir Hugh Graham, president of the Montreal Star, whose name figured so prominently in the 1917–19 debate on honours. This appointment was by far the most controversial, for Graham was remembered vividly for his fundraising activities on behalf of the Conservatives as well as his stridently pro-imperial editorials in the Star at the time of the South African War (1899–1902). The appointment — made over Borden's objections and politically ill-timed — may well have been the turning point in rendering the prospect of Canadian peers a politically unpalatable one.
There have been a number of Canadians appointed to the peerage of the United Kingdom for services to Britain, and these are therefore not considered "Canadian" peerages. Sir Arthur Lawrence Haliburton, 1st Baron Haliburton, was born in Nova Scotia and rose to become the permanent undersecretary of state for war in Britain from 1895 to 1897. William James Pirrie, head of the famous Belfast shipbuilding firm Harland and Wolff, was made a baron in 1906 and later elevated as Viscount Pirrie in 1921. Sir Max Aitken, a native of New Brunswick, was raised to the peerage as 1st Baron Beaverbrook in 1917, yet another Canadian newspaper owner sent to the House of Lords. Sir Edward Patrick Morris, the premier of Newfoundland, became the island's only native son to be made a peer, as 1st Baron Morris of St. John's, in 1918. Although Newfoundland was not part of Canada at the time, it seems appropriate that Morris should be included in this list. Richard Bedford Bennett, the Canadian prime minister who revived the imperial honours system in Canada during the 1930s, was raised to the peerage as 1st Viscount Bennett in 1941. This occurred several years after Bennett had moved to Britain. Following the long tradition of Canadian newspapermen being ennobled, Roy Thomson was made 1st Baron Thomson of Fleet in 1963. In 2000, Conrad Black became Lord Black of Crossharbour and is the only Canadian yet to be appointed a life peer.
During the 1902 coronation of King Edward VII, the new King was keen to make Sir Wilfrid Laurier "Baron Laurier of St. Lin." Laurier was shocked at the offer and declined. In Canada peerages have historically been highly contentious and continue to be as evidenced by the case of Lord Black of Crossharbour, who was forced to renounce his Canadian citizenship in order to become a peer. Upon his retirement as governor general of Canada, Lord Alexander was raised in the peerage from a viscount to an earl and given the additional title of Baron Rideau of Ottawa and of Castle Derg, County Tyrone.
Number of Elevations: In total, ten Canadians were elevated to the peerage from 1867 until the present day; only the first five can be considered Canadian appointments:
* • Lady Susan Agnes Macdonald, Baroness Macdonald of Earnscliffe in the Province of Ontario and of the Dominion of Canada (1891).
* • Sir George Stephen, Bt, GCVO, 1st Baron Mount Stephen of Mount Stephen in the Province of British Columbia and Dominion of Canada, and of Dufftown in the county of Banff, New Brunswick (1891).
* • Sir Donald Alexander Smith, GCMG, GCVO, 1st Baron Strathcona and Mount Royal (1900).
* • Sir Thomas Shaughnessy, Kt, KCVO, 1st Baron Shaughnessy of the City of Montreal in the Dominion of Canada and Ashford in the County of Limerick (1916).
* • Sir Hugh Graham, Kt, 1st Baron Atholstan of Huntington in the Province of Quebec in the Dominion of Canada and of the City of Edinburgh (1917).
* • Sir William Maxwell Aitken, Kt, 1st Baron Beaverbrook in the Province of New Brunswick in the Dominion of Canada and Cherkley in the County of Surrey (1917); posthumously made an ONB.
* • Sir William James Pirrie, KP, 1st Baron Pirrie of the City of Belfast (1906). Elevated to 1st Viscount Pirrie of the City of Belfast (1921).
* • The Right Honourable Richard Bedford Bennett, PC, KC, ED, 1st Viscount Bennett of Mickelham, Calgary, and Hopewell (1941).
* • Roy Herbert Thomson, 1st Baron Thomson of Fleet (1964), subsequently made a GBE.
* • The Right Honourable Conrad Black, Baron Black of Crossharbour (1999).[1]
Unlike the other honours covered throughout this book, there are no distinctive insignia to indicate a peerage. Instead, peers are entitled to wear a coronet and robe appropriate to their degree at the coronation of the sovereign (the coronet can also be displayed with the peer's coat of arms), and for those in the House of Lords, a special parliamentary robe can be worn on select occasions.
The Baronetage in the United Kingdom
Baronet of the United Kingdom insignia.
Origins: Baronets rank immediately after barons but are not part of the peerage and do not have the right to sit in the House of Lords. A baronetcy is somewhat like a hereditary knighthood, taking the same title as a knight but without the dubbing. The title is passed down like a peerage. The honourable order of baronets consists of five divisions:
> I Baronets of England, 1611
>
> II Baronets of Ireland, 1619
>
> III Baronets of Scotland, 1625 (Baronets of Nova Scotia)
>
> IV Baronets of Great Britain, 1707
>
> V Baronets of the United Kingdom, 1801
Since 1801 all baronetcies created have been baronetcies of the United Kingdom.
In total, seven baronetcies were awarded upon the recommendation of the Canadian government: three were awarded to politicians and four to businessmen. The last baronet of the United Kingdom was appointed in 1991 when Dennis Thatcher, husband of Margaret Thatcher, was appointed 1st Baronet Thatcher of Scotney in the County of Kent. The last Canadian appointed as a baronet for services to Canada was Sir Joseph Flavelle in 1917.
As the honour is hereditary, there remain hundreds of baronets despite the fact that no creations have been made for more than twenty years. There continue to be Canadian holders of these honours, since they are hereditary.
Criteria: Service national in scope.
Insignia (for all except baronets of Nova Scotia): A gold oval badge 55 mm in length and 30 mm in width, the centre bearing a single escutcheon displaying the red hand of Ulster surmounted by a Tudor Crown. The circumference of the insignia is decorated with roses (for baronets of England), thistles (of Scotland), roses and thistles (of Great Britain) or roses, thistles and shamrocks (of the United Kingdom) in filigree on a dark blue enamel background. The reverse is plain, usually with four or six rivet heads and the appropriate gold or silver hallmarks. Occasionally, the reverse will be engraved with the details of the recipient.
Suspender: A straight gold loop suspender.
Ribbon: A yellow ribbon 44 mm wide edged on each side with 4 mm of royal blue.
Postnominals: Bt or Bart.
Other: The insignia is found in both sterling gilt and carat gold.
Number of Creations for Canadians:7.
The Baronetage of Nova Scotia
The baronets of Nova Scotia are addressed separately, since the creation of baronets of Nova Scotia originally pertained directly to the colonization of what are today the provinces of Nova Scotia and New Brunswick.
Origins: Created as part of Sir William Alexander's scheme to prevent Nova Scotia from falling into French hands by settling the territory, the baronets of Nova Scotia were established by King James I as a way to raise money independently of Parliament. Charles I granted the holders of this land the title "Sir" and allowed for the creation of an insignia for the baronets. Baronets of Nova Scotia ceased to be created in 1707 afer the union of Scotland and England into Great Britain. Initially, the baronets were entitled to have a hereditary seat in the provincial legislative assembly (not yet established) and were imbued with the authority of a feudal baron upon their own lands.
Baronet of Nova Scotia insignia.
Criteria: Service to the Crown of Scotland and/or purchase of a grant of land in the Royal Province of Nova Scotia.
Insignia: A gold oval insignia varying in height from 54 mm to 41 mm. The centre displays the shield of Nova Scotia surmounted by a crown. The circumference of the insignia displays the text FAX MENTIS HONESTAE GLORIA ("Glory is the light of a noble mind") in gold on a field of blue enamel. The reverse is plain, though it is occasionally engraved with the details of the recipients and gold or silver hallmarks.
Suspender: A gold loop suspender of varying sizes.
Ribbon: A watered orange tawny ribbon 44 mm in width.
Postnominals: Bt, Bart, or Bart NS.
Other: A variety of manufacturers have produced the insignia. Some pieces in carat gold date from the founding of the baronetage and are quite rare.
Number of Creations: 329.
The Most Honourable Order of the Bath
The Order of the Bath was founded by King George I in 1725 as a single-level military order consisting of Knights of the Bath (KB). In 1815 a civil division was added to the order and the military division was restructured into three levels: Knight Grand Cross (GCB), Knight Commander (KCB), and Companion (CB). By 1847 the civil division was restructured to mirror its military counterpart and was divided into three levels.[2] Appointments to the military division were made in recognition of service of the highest calibre, and nominees had to have a Mention in Dispatches. Those appointed to the order generally started at the CB level and could subsequently be elevated to KCB and GCB. Officers holding the rank of major or commander and above were eligible for appointment to the order, with KCBs being awarded primarily to those holding the rank of major-general and above, and GCBs being bestowed upon lieutenant-generals, generals, and field marshals or their naval equivalents.
No member of the Canadian Militia was ever appointed a GCB, though Canadians serving in the British Army did receive this honour, notably General Sir William Fenwick Williams and General Sir Gordon Drummond. In the pre-Confederation period the only Canadian appointed to the military division of the Order of the Bath was Lieutenant-Colonel Charles de Salaberry in 1817 for services in the War of 1812 at the Battle of Chateauguay. Eight members of the Canadian Militia were made KCBs, General Sir Arthur Currie and Victoria Cross holder Lieutenant-General Richard Turner being the most prominent, both in recognition of services rendered during the First World War.
During the South African War (1899–1902), five Canadians were made Companions; the First World War saw forty-two Canadians appointed CBs; two were made during the interwar period; sixty-two during the Second World War; and one final appointment went to Brigadier John Rockingham for service in the Korean War. The final Canadian KCB in the military division came during the brief Bennett honours revival of 1933–35 when the commissioner of the RCMP, Major-General James Howden MacBrien, was appointed.
In the pre-Confederation period, seven Canadians were appointed to the civil division as Companions of the Order of the Bath. Immediately following Confederation, the civil division of the Order of the Bath was the principal means through which the Crown honoured senior office-holders and politicians. The most famous of those appointed to the Order of the Bath was Sir John A. Macdonald, who was made a KCB upon the achievement of Confederation. He would go on to be made a GCB in 1884. Seven Fathers of Confederation were made Companions of the Order of the Bath in recognition of their contributions, with one refusing the honour. George- Étienne Cartier declined appointment as a CB in 1867 because he felt he deserved an honour equal to that accorded to his English-Canadian counterpart, Macdonald, as the representative of French Canada. Cartier's patience paid off, and in 1868 he was appointed a baronet, an honour superior to Macdonald's KCB!
In 1868 the statutes of the Order of St. Michael and St. George were altered and services that had previously been recognized with the Order of the Bath in Britain's overseas dominions and colonies were henceforth recognized with the Order of St. Michael and St. George (the civil division of the Order of the Bath continued to be used to honour senior civil servants within the United Kingdom). This explains why Canadian appointments to the civil division of the order were nearly discontinued after 1868. Sir Sam Hughes and Major-General Sir John Wallace Carson were the last Canadians to be appointed to the civil division of the order, both as KCBs.
Criteria: Exemplary service at the national level in either military or civil service.
Insignia: Distinct insignia exist for the military and civil divisions.
Ribbon: Crimson, 38 mm wide for all levels, save the GCB sash, which is 102 mm wide.
Knight Grand Cross (Military and Civil Divisions)
Order of the Bath appointment warrant.
Mantle: Crimson satin, lined with white taffeta and tied with a cord of white silk, to the end of each which is attached crimson silk and gold tassels. The breast star of a Knight Grand Cross in the military or civil division is embroidered in wire on the left side of the mantle.
Collar: The collar insignia for the order is the same for both the military and civil divisions. The sash badge is attached to the collar; the collar badges differ for the civil and military divisions. A collar of gold composed of nine arched Imperial Crowns, eight roses, thistles, and shamrocks issuing from a gold sceptre enamelled in their proper colours, linked together with seventeen gold-enamelled knots.
Postnominals: GCB.
Knight Grand Cross (Military Division)
Sash Badge: A gold 88 mm wide Maltese cross with eight points, enamelled white, edged in gold, in each of the four angles displaying a lion passant guardant, in the centre on a gold background, a rose, thistle, and shamrock issuing from a sceptre, between three arched Imperial Crowns, this central device contained within a gold border displaying the motto TRIA JUNCTA IN UNO ("Three joined into one") encompassed by two laurel branches in green with a blue ribbon at the base containing the text ICH DIEN ("I serve"). The sash is worn over the right shoulder with the sash badge resting on the left hip.
Knight Grand Cross of the Order of the Bath (military).
Breast Star: A star 88 mm wide and 95 mm in height composed of silver rays, displaying in the centre a Maltese cross in gold, the centre of which is gold upon which there are displayed three arched Imperial Crowns in gold, the central device circumscribed by a gold ribbon containing the motto TRIA JUNCTA IN UNO and bordered on each side one laurel branch in green, at the base a small blue scroll with the text ICH DIEN.
Knight Grand Cross (Civil Division)
Sash Badge: A gold oval badge 42 mm wide, displaying a rose, thistle, and shamrock emanating from a sceptre between three arched Imperial Crowns, the whole of the centre pierced and circumscribed by the motto TRIA JUNCTA IN UNO.
Knight Grand Cross of the Order of the Bath (civil).
Breast Star: A star 88 mm wide and 95 mm in height composed of silver rays, the centre of which is gold upon which there are displayed three arched Imperial Crowns in gold, the central device circumscribed by a gold ribbon containing the motto TRIA JUNCTA IN UNO. The sash is worn over the right shoulder with the sash badge resting on the left hip.
Knight Commander (Military Division)
Knight Commander of the Order of the Bath (military).
Neck Insignia: Similar to the sash badge worn by GCBs in the military division, though 58 mm wide. The suspender ring through which the ribbon passes displays oak leaves.
Breast Star: Composed of four rays of silver, 76 mm wide and tall, between each of which issues a smaller ray of silver, displaying in the centre a Maltese cross in gold, the centre of which is gold upon which there are displayed three arched Imperial Crowns in gold, the central device circumscribed by a gold ribbon containing the motto TRIA JUNCTA IN UNO and bordered on each side one laurel branches in green, at the base a small blue scroll with the text ICH DIEN.
Postnominals: KCB.
Knight Commander (Civil Division)
Knight Commander of the Order of the Bath (civil).
Neck Insignia: Similar to the sash badge worn by GCBs in the civil division, though 38 mm wide. The suspender ring through which the ribbon passes is found both ribbed and plain.
Breast Star: Composed of four rays of silver, 76 mm in width, between each issues a smaller silver ray, in the centre on a gold background, three arched Imperial Crowns, the centre device circumscribed by a circlet of red displays the motto TRIA JUNCTA IN UNO.
Postnominals: KCB.
Companion (Military Division)
Companion of the Order of the Bath (military) on buckle.
Insignia: Similar to the neck insignia worn by a Knight Commander in the military division, though 52 mm wide. No breast star is worn by Companions.
Suspender: A plain or ribbed gold ring through which the ribbon passes. Until 1917, Companions in both divisions wore their insignia on a straight suspender and gold buckle on the left breast.
Postnominals: CB.
Companion of the Order of the Bath (military) neck issue.
Companion (Civil Division)
Companion of the Order of the Bath (civil) on buckle.
Companion of the Order of the Bath (civil) neck issue.
Neck Insignia: Similar to the neck insignia worn by a Knight Commander in the civil division, though it is 32 mm wide. No breast star is worn by Companions.
Suspender: Same as for the CB military division. Until 1917, Companions in both divisions wore their insignia on a straight suspender and gold buckle on the left breast.
Postnominals: CB.
Insignia Manufacture: Insignia presented to Canadians from 1867 to 1946 were manufactured by Garrard & Co.
Number of Appointments of Canadians: 141.
* • GCB Military: 0.
* • GCB Civil: 1.
* • KCB Military: 8.
* • KCB Civil: 3.
* • CB Military: 113.
* • CB Civil: 16.
The Most Distinguished Order of St. Michael and St. George
Created by King George III in 1818 to recognize services in the Ionian Islands, the Order of St. Michael and St. George was restructured and given a new purpose in 1868 when it was then converted into an order to recognize service in the foreign and colonial services.
From Confederation until the Nickle Resolution, it can be said that this order served as the era's "Order of Canada," as it was the premier award available to Canadian politicians, judges, and civil servants. This was partly because of a change in the rules in 1868, which tipped the scales away from the Order of the Bath and toward this order. When first established, the Order consisted of the Sovereign, grand master, seven Knights Grand Cross, twelve Knight Commanders, and twenty-four Knight Companions. In 1832 the lowest class became non-titular and was renamed Companion of the Order of St. Michael and St. George. The order was enlarged in 1868, 1877, and 1902 to allow for a maximum of 110 Grand Crosses (GCMG), 390 Knight Commanders (KCMG), and 1,775 Companions of the Order of St. Michael and St. George (CMG). For Canadians the order was intended to honour "high political or official services as distinct from judicial, literary, scientific, or artistic eminence."
From 1868 until 1935, fourteen GCMGs (the award was reserved for governors general, prime ministers, and chief justices) were awarded to Canadians, including three Canadian prime ministers: Sir Charles Tupper, Sir Wilfrid Laurier, and Sir Robert Borden. Many other Canadians keenly sought to become KCMGs, partly because it was the most accessible honour within the reach of eminent senior civil servants, senators, cabinet ministers, provincial premiers, or even wealthy party supporters, and perhaps also because of the splendid insignia involved. Seventy-nine of them were successful: thirty-six politicians, sixteen serving or retired lieutenant governors — who were, in general, former politicians — while the remainder went to people in a variety of occupations ranging from university administrators to engineers.
The CMG was bestowed upon 370 Canadians. Even William Lyon Mackenzie King was awarded this distinction. The last Canadian appointments to the order came in 1946 when forty appointments were made at the level of CMG. This last batch was certainly the most diverse, with such prominent Canadians as the Group of Seven painter A.Y. Jackson and the poet Edwin John Pratt being recognized alongside a variety of deputy ministers and leaders in the transportation industries.
Criteria: Exemplary service in the foreign, dominion, or colonial services.
Insignia: Varies depending on level of appointment.
Ribbon: A watered ribbon of Saxon blue divided in the centre by an equal stripe of red; 38 mm for all levels, save the GCMG sash, which is 102 mm wide.
Knight Grand Cross
Knight Grand Cross of the Order of St. Michael and St. George (part of collar also shown.).
Mantle: Saxon blue lined with crimson silk and tied with two cords of blue and crimson silk and gold. The breast star of a Knight Grand Cross is embroidered in wire on the left. There is also a hat of blue satin lined with scarlet and finished with black and white ostrich feathers.
Collar: Gold in colour composed alternatively of twenty-four devices: lions of England crowned, en-amelled Maltese crosses, and the cyphers SM and SG, the centre of the collar having an Imperial Crown flanked on each side by a winged lion passant guardant, each holding in his forepaw a book of seven arrows, At the back where the two halves of the chain are joined there are two similar lions.
Sash Insignia: A fourteen-point gold cross 78 mm wide, enamelled in white with gold edging the entire insignia surmounted by a crown. The centre of the obverse displays a representation of St. Michael trampling on Satan, the entire central device circumscribed by a blue ribbon containing the motto AUSPICIUM MELIORIS ÆVI ("Token of a better age"). The reverse is identical to the obverse, though the centre displays a representation of St. George and the dragon. The sash is worn over the right shoulder with the sash badge resting on the left hip.
Breast Star: A silver breast star composed of seven rays, each ray having in its centre a single gold ray. Superimposed on this is a red enamelled Cross of St. George, edged in gold, the centre displaying a representation of St. Michael trampling on Satan, the entire central device circumscribed by a blue ribbon containing the motto AUSPICIUM MELIORIS ÆVI.
Postnominals: GCMG.
Knight Commander
Knight Commander of the Order of St. Michael and St. George.
Neck Insignia: The same as the Knight Grand Cross sash insignia, though 52 mm wide. The ribbon passes through a ribbed gold ring.
Breast Star: A silver breast star composed of four rays surmounted by a Maltese cross in saltire. Superimposed on this is a red enamelled Cross of St. George, edged in gold, the centre displaying a representation of St. Michael trampling on Satan, the entire central device circumscribed by a blue ribbon containing the motto AUSPICIUM MELIORIS ÆVI.
Postnominals: KCMG.
Companion
Companion of the Order of St. Michael and St. George on buckle.
Neck Insignia: Identical to that of a Knight Commander, though 45 mm in width.
Suspender: The ribbon passes through a ribbed gold ring. Until 1917, Companions in both divisions wore their insignia on a straight suspender and gold buckle on the left breast.
Postnominals: CMG.
Insignia Manufacture: Insignia awarded to Canadians were made by Garrard & Co.
Companion of the Order of St. Michael and St. George neck issue.
Number of Appointments of Canadians: 463.
* • GCMG: 14.
* • KCMG: 79.
* • CMG: 370.
Order of St. Michael and St. George appointment warrant.
The Most Excellent Order of the British Empire
The Order of the British Empire was founded on June 4, 1917, as an honour initially intended to recognize the effort of civilians and the military who were serving away from the front. During the Second World War, it served as Canada's national order, being presented for both civil and military achievements. The order was established by King George V as a way to recognize ordinary people throughout the British Empire for a diverse array of contributions both civil and military and today is the most widely bestowed honour in the United Kingdom. It was the first modern honour in that it consisted of five levels, was not limited to the elite, and women were admitted from the time of its establishment.
The other imperial orders of chivalry were primarily reserved for those serving in the senior levels of the military, civil service, or politics, so the Order of the British Empire was created as an order to recognize everyone from the most junior war workers and volunteers right up to the great men and women of the day. Like the Royal Victorian Order, the Order of the British Empire was established with five levels, similar to France's Légion d'honneur. It also initially included the Medal of the Order of the British Empire; however, this was converted into the British Empire Medal in 1922.
Crossed oak leaves used on the ribbon of the Order of the British Empire (1957—74) to denote gallantry.
The first Canadian to be invested with the order was the Montreal manufacturer Sir Charles Blair Gordon, who had served for two years as the chairman of the Imperial Munitions Board. He was appointed a Knight Commander of the Order of the British Empire (KBE) and invested by King George V on October 18, 1917.
As prime minister, R.B. Bennett drew up an honours list with 163 Canadians to be appointed to the Order of the British Empire. It included more women and people from the volunteer sector than had ever been recognized with honours in the history of the empire. Bennett saw the order as an effective tool to recognize national and local contributions that brought benefit to other citizens from all walks of life. In many ways it was Bennett's conception of honours for "exemplary citizenship" that was used as the basis behind the modern Canadian honours system. Some of the most notable Canadians appointed to the order include Sir Edward Beatty, president of the CPR, who was made a GBE in 1935; Sir Frederick Banting, the co-discoverer of insulin, who was made a KBE in 1934; and author Lucy Maud Montgomery and diplomat (and future prime minister) Lester Pearson, who were made CBE and OBE respectively.
Criteria: Exemplary service at the international, national, or local levels that brought credit to the British Empire or one of its component parts. The order was also occasionally used to recognize gallantry, and from 1957 to 1974, such awards were noted by the wearing of a pair of crossed oak leaves on the ribbon. The most notable example was General Jacques Dextraze who, as a brigadier, was awarded the CBE for gallantry during the U.N. Mission to the Congo in 1964.
Insignia: Varies according to level of appointment. The reverse of all neck and breast insignia displays a rope circlet and the Royal Cypher of King George V, GRI, surmounted by a Tudor Crown. Some OBE and MBE insignia also carry sterling hallmarks. The 2nd-type insignia are described below; the 1st-type insignia differed in that the central medallion displayed a seated effigy of Britannia holding a trident in place of the conjoined crowned busts of King George V and Queen Mary, the 2nd type having been introduced in 1937. The other difference between the 1st and 2nd types applied to the GBE and KBE breast stars, which had smooth and not chipped rays.
Ribbon: There have been four types of ribbon; two for the civil division and two for the military division.
* • The sash for all GBEs is 102 mm wide; the ribbon for all other levels of the order is 38 mm wide; while the BEM has a ribbon 32 mm in width.
* • 1st-type civil (1917–1937) purple.
* • 1st-type military (1918–1937) purple with a central 3 mm strip of scarlet.
* • 2nd-type civil (1937 to present) rose pink edged with 3 mm of grey on each side.
* • 2nd-type military (1937 to present) rose pink edged with 3 mm of grey on each side and a central 1 mm stripe of grey.
Knights/Dames Grand Cross
Knight Grand Cross of the Order of the British Empire 1st type (civil ribbon).
Mantle: Rose pink satin, lined with pearl grey silk, fastened with a cord of pearl grey silk, having two rose pink and silver tassels. A breast star of a Knight/Dame Grand Cross is embroidered in wire on the left side.
Collar: A gold collar made up of six medallions of the Royal Arms alternating with six medallions displaying the Royal Cypher of King George V, linked together by cables thereon an Imperial Crown between two sea lions, each holding a trident.
Sash Insignia: A gold cross patonce 70 mm wide, translucent pearl light blue enamel in colour with machine work hatching beneath the enamel, in the centre a gold medallion bearing the conjoined crowned effigies of King George V and Queen Mary, circumscribed by a maroon enamel border and the words FOR GOD AND THE EMPIRE. The uppermost arm of the cross is surmounted by an Imperial Crown in gold. The sash is worn over the right shoulder with the sash badge resting on the left hip.
Breast Star: An eight-pointed star 86 mm by 115 mm with chipped silver rays, in the centre of which is a gold medallion bearing the conjoined crowned effigies of King George V and Queen Mary, circumscribed by a maroon enamel border and the words FOR GOD AND THE EMPIRE.
Postnominals: GBE.
Knight/Dame Commander
Knight Commander of the Order of the British Empire 1st type (military ribbon).
Neck Insignia: Identical to that of a Knight/Dame Grand Cross, though 64 mm in width. The ribbon passes through a plain gold ring.
Breast Star: A four-pointed star 80 mm wide by 80 mm high with chipped silver rays, in the centre a gold medallion bearing the conjoined crowned effigies of King George V and Queen Mary, circumscribed by a maroon enamel border and the words FOR GOD AND THE EMPIRE.
Knight Commander of the Order of the British Empire 2nd type (breast star).
Postnominals: KBE or DBE.
Commander
Commander of the Order of the British Empire 1st type (civil ribbon).
Commander of the Order of the British Empire 2nd type (civil ribbon).
Insignia: Identical to that worn by Knights Commander/Dames Commander of the order. Commanders do not wear a breast star.
Suspender: The ring passes through a plain gold ring attached to the top of the insignia.
Postnominals: CBE.
Officer
Officer of the Order of the British Empire 2nd type (civil ribbon).
Officer of the Order of the British Empire 1st type (civil ribbon).
Insignia: Similar to that of a Commander, though 51 mm in width and rendered in gold without enamel. Beginning in the early 2000s, the insignia were changed from a matte to a polished bright gold appearance.
Suspender: A plain gold ring attached to the top of the insignia.
Postnominals: OBE.
Member
Member of the Order of the British Empire 1st type (military ribbon).
Member of the Order of the British Empire (civil ribbon) reverse.
Member of the Order of the British Empire 2nd type (military ribbon).
Insignia: Identical to that worn by officers, though in plain frosted silver. Beginning in the early 2000s, the insignia were changed to a polished bright silver appearance.
Suspender: A plain gold ring attached to the top of the insignia.
Postnominals: MBE.
Insignia Manufacture: The insignia for the GBEs, KBEs/DBEs, and CBEs awarded to Canadians were all made by Garrard & Co. The insignia for the OBE and MBE 1st type were also made by Garrard & Co.; the OBE and MBE of the 2nd type were made by the Royal Mint.
Number of Appointments of Canadians: 4,753.
* • GBE: 2 (both civil division 1st type).
* • KBE: 8 (all civil division 1st type).
* • CBE: 427.
* ɣ 4 (civil division 1st type).
* ɣ 52 (military division 1st type).
* ɣ 160 (civil division 2nd type).
* ɣ 211 (military division 2nd type).
* • OBE: 1,643.
* ɣ 59 (civil division 1st type).
* ɣ 261 (military division 1st type).
* ɣ 533 (civil division 2nd type).
* ɣ 781 (military division 2nd type).
* • MBE: 2,673.
* ɣ 79 (civil division 1st type).
* ɣ 108 (military division 1st type).
* ɣ 743 (civil division 2nd type).
* ɣ 1,734 (military division 2nd type).
Medal of the Order of the British Empire
Created at the same time as the Order of the British Empire in 1917, the Medal of the Order of the British Empire was meant to be a junior award for labourers and workers who were involved in wartime endeavours. The medal was divided into military and civil divisions in December 1918. Canadians were awarded the medal between 1918 and 1920, with all awards being made in recognition of courageous acts rendered in industrial settings. The medal was phased out in 1922 and replaced by the British Empire Medal and the Empire Gallantry Medal (see chapter 5).
Criteria: The medal was awarded for services of special merit rendered by men and women in manual and other work done in connection with the First World War. Such service included acts of great courage, self-sacrifice, initiative, perseverance, skill, invention, or resource.
Medal of the Order of the British Empire (civil ribbon) obverse.
Insignia: A circular silver medal 27 mm in diameter, the obverse depicting a seated effigy of Britannia holding a trident circumscribed by the text FOR GOD AND THE EMPIRE, the reverse bearing the Royal Cypher GRI surmounted by an Imperial Crown and surrounded by a rope motif.
Suspender: A silver ring 15 mm in diameter through which the ribbon passes.
Ribbon: Crimson in colour and 27 mm wide for the civil division. A central 2 mm stripe of scarlet was added for the military division.
Naming: None.
Postnominals: None.
Other: The medals were issued unnamed and were struck by John Pinches.
Number: 44 total: one military, 43 civil.
British Empire Medal
The British Empire Medal was created by King George V in 1922 to replace the Medal of the Order of the British Empire. While originally styled "The Medal of the Order of the British Empire (for Meritorious Service)," this was superseded by the more succinctly titled British Empire Medal, a designation that became official in 1941. Prior to the establishment of the Canadian decorations for bravery, the BEM was also used in Canada as a third-tier gallantry award, with the last BEM for gallantry being awarded to a Canadian in 1968.
British Empire Medal (military ribbon) obverse.
British Empire Medal (military ribbon) reverse.
Criteria: Initially instituted to recognize meritorious service rendered by non-commissioned officers and men who were not eligible for appointment to the military division of the Order of the British Empire, the civil BEM was used to recognize public servants and people involved in jobs below management posts. The military and civil BEMs were also used to recognize acts of gallantry.
Insignia: A circular silver medal 36 mm in diameter, the obverse depicting a seated effigy of Britannia holding a trident and circumscribed by the text FOR GOD AND THE EMPIRE with FOR MERITORIOUS SERVICE at the base. The reverse displays the Royal Cypher surmounted by a crown in the centre (GRIV) circumscribed by four lions. The reverse of the medal was changed following the death of King George V and has subsequently displayed the Royal Cypher (GRI, GVIR, or EIIR) surmounted by a crown in the centre and circumscribed by two lions on each side with the text INSTITUTED BY KING GEORGE V.
Suspender: A claw suspender composed of a trio of stylized oak leaves attached to a straight suspension bar.
Ribbon: Same variations as the ribbon of the Order of the British Empire, though 32 mm in width.
Bars: The BEM could be awarded with a bar; however, no Canadian received one. From 1957 to 1974, those awarded the BEM for gallantry were entitled to wear silver oak leaf devices on their ribbons to indicate that their awards were made in recognition of gallant acts.
Naming: Engraved or impressed capitals around the rim.
Postnominals: BEM.
Other: Struck at the Royal Mint.
Number Awarded to Canadians: 1,318.
* • Military George V issue: 2.
* • Civil George V issue: 1.
* • Military George VI issue: 1,143 (Second World War).
* • Civil George VI issue: 97 (Second World War).
* • Military George VI issue: 38 (27 Korean War + 11 post-Korea).
* • Civil George VI issue: 0.
* • Military Elizabeth II issue: 15.
* • Civil Elizabeth II issue: 22 (all gallantry).
The Order of the Companions of Honour
Founded by King George V on June 4, 1917, the order was originally established as a non-titular alternative to the various imperial orders of chivalry. Over time it has largely been treated as a junior level of the Order of Merit. Originally limited to only fifty members, it was initially bestowed upon leaders who had performed conspicuous service during the First World War.
In 1943 the British government decided to expand the order, and Canada, Australia, New Zealand, and South Africa were asked whether or not they would like a certain number allotted for their residents. Australia and New Zealand were keen to be involved while, not surprisingly, Canada and South Africa declined the offer. At this time Canadian officials were in the process of creating the Canadian Award of Honour, the Canadian Decoration of Honour, and the Canada Medal. The Canadian government felt that with the impending establishment of these new Canadian awards there would be no need for Britain to put aside a certain number of CH appointments.
Order of the Companions of Honour obverse.
The revised 1943 statutes of the order set out the following quotas: United Kingdom, forty-five; Australia, seven; New Zealand, two; other Commonwealth nations, eleven — for a total of sixty-five. Had Canada accepted the British offer, it would have been allotted seven appointments, while Australia's quota would have been reduced to five. Appointments would have been made by the Sovereign based on the advice of the Canadian prime minister, a procedure that was used in the United Kingdom, Australia, and New Zealand (upon the advice of their respective prime ministers).
The first CH to be bestowed upon a Canadian was awarded to General Harry Crerar, the general officer commanding the 1st Canadian Army in 1944. He had been nominated as a Knight Commander of the Order of the British Empire, but the Canadian government would not entertain such an award. The CH was therefore considered because it was not a knighthood, yet it was still a senior honour. The recommendation to King George VI came from the Canadian minister of national defence after being cleared by the prime minister.
Order of the Companions of Honour reverse.
Vincent Massey's CH appointment, the second to a Canadian, was initiated by King George VI. Initially, Prime Minister Mackenzie King refused to approve the appointment, but he acquiesced after considering Massey's outstanding record of service as high commissioner to London.
General Andrew McNaughton was appointed to the order on the advice of the Canadian prime minister. He was the last Canadian to be appointed to the order for twenty-five years. This dry spell ended when Dr. Charles Best, the co-discoverer of insulin, was appointed in 1971. He was followed by Arnold Smith, the first secretary general of the Commonwealth, who was appointed to the CH in 1975 at the initiative of Queen Elizabeth II.
John Diefenbaker had refused the order when he was first offered it in 1962 — at which time he was serving as prime minister — but accepted an appointment in 1976. The British prime minister put his name forward after consulting Prime Minister Pierre Trudeau. There is some evidence that Trudeau initiated the appointment, feeling that Diefenbaker deserved recognition and no doubt being aware that, as an active politician, he was ineligible for the Order of Canada. Trudeau himself would be made a CH in 1984, shortly after his final retirement from politics. His appointment was initiated by both the British and Canadian governments.
A recent Canadian to be appointed to the order is General John de Chastelain, former chief of the defence staff and former Canadian ambassador to the United States. His appointment was in recognition of his work on the Northern Ireland peace process. General de Chastelain's award is considered to be a "British" one because his nomination was made by the British prime minister for work in the United Kingdom. Similarly, the appointment of Professor Anthony Pawson for his work on cell behaviour was made on the advice of the British government.
Criteria: Any citizen of the Commonwealth who has rendered conspicuous service of national importance. Like the Order of Merit, in the Canadian context it has been used to recognize outstanding service, which in other parts of the Commonwealth would usually be recognized through the bestowal of a knighthood.
Order of the Companions of Honour appointment warrant.
Insignia: A silver gilt oval badge 75 mm tall and 42 mm wide surmounted by a Tudor Crown or Royal Crown. The circumference of the badge is enamelled in blue and bears the motto IN ACTION FAITHFUL AND IN HONOUR CLEAR. At the centre of the insignia is a rectangular plaque containing an effigy of a knight on horseback under an oak tree. The reverse of the insignia carries the Royal Cypher surmounted by a crown. It was not until 1988 that the crown on the insignia was changed to a St. Edward's Crown, despite the fact that the Queen had requested the change in 1953.
Suspension: A small ring passes through a mounting loop, which is mounted on the reverse of the crown. Through this small teardrop-shaped ring passes a decorative loop, through which the ribbon passes.
Ribbon: Carmine in colour, with borders of gold thread, 40 mm in width.
Postnominals: CH.
Other: This insignia is not worn in miniature.
Insignia Manufacture: Insignia presented to Canadians were made by John Pinches and most recently by Toye, Kenning & Spencer Ltd.
Number of Canadian Appointments: 9
* • General Harry Crerar, CH, CB, DSO, CD (1945).
* • The Right Honourable Vincent Massey, PC, CH, CC, CD (1946).
* • General the Honourable Andrew George Latta McNaughton, PC, CH, CB, CMG, DSO, CD (1946).
* • Dr. Charles Best, CC, CH, CBE (1971).
* • Arnold Smith, OC, CH (1975).
* • The Right Honourable John George Diefenbaker, PC, CH, QC (1976).
* • The Right Honourable Joseph Philippe Pierre Yves Elliott Trudeau, PC, CC, CH, QC (1984).
* • General Alfred John Gardyne Drummond de Chastelain, CC, CMM, CD, CH (1998).
* • Professor Anthony Pawson, OC, OOnt, CH (2006).
Knights Bachelor
Knights Bachelor insignia.
Knights of the United Kingdom (also known as Knights Bachelor because of the title's non-hereditary character) comprise the most junior form of knighthood. Established in the eleventh century, the honour has become syn-onymous with chivalry and service. Knights Bachelor are permitted to prefix their names with the appellation "Sir." Women, though, have never been granted this honour. Since 1917, women have been appointed as Dames of the Order of the British Empire instead.
A knighthood is a title conferred by the Sovereign, and Knights Bachelor are not connected with any order: they are members of "the ancient concept of knighthood and chivalry" and become such when the Sovereign lays the blade of his or her sword upon the right and then the left shoulder of a kneeling subject. Beginning in 1926, Knights Bachelor were presented with an oval breast insignia worn in the same fashion as a breast star. In 1973 this was replaced with an oval neck badge, and since 1988 recipients are permitted to wear both the oval neck badge and the breast insignia.
Most Canadians who were knighted became Knights Bachelor, and 107 received the honour for services rendered in and to Canada. A total of nine were awarded to Newfoundlanders prior to that dominion joining Canadian Confederation.
The award was limited, but not always confined, to the bench and men of literary or scientific note. Pressure to award a knighthood to the politically connected came from Canada. Provincial and puisne judges and chief justices were most frequently given the honour, receiving forty-two in total, while politicians received eighteen and businessmen fifteen.
Criteria: Service national in scope in a variety of fields.
Insignia: Three types of insignia have been issued for this order, though it is only the 2nd type that was awarded to Canadians. The 1st type (1926–33) is a gold oval-shaped breast badge 76 mm high and 56 mm wide, displaying within a scroll, a cross hilted sword belted and sheathed, pommel upward between two spurs, rowels upward, the whole set about with the sword belt. The 2nd type (1933–73) is identical to the 1st, though smaller in size, measuring 56 mm high and 32 mm wide. The 3rd type (1973 to present) is identical to the 1st type, 50 mm high and 36 mm wide, hung from a loop on a ribbon around the neck.
Suspender: For the breast insignia, a narrow gold pin and clasp on the reverse of the insignia for attaching to the jacket or uniform tunic. For the neck insignia from 1973, an oblong wire loop 18 mm tall.
Ribbon: Introduced in 1973, crimson in colour 38 mm wide bordered on each side by a 3 mm of yellow. The only Canadian Knights Bachelors living at the time were Sir Ernest MacMillan, Sir Leonard Outerbridge, and Sir William Stephenson.
Postnominals: Kt, almost solely used in Canada.
Other: The insignia were made by a number of manu-facturers, notably the Royal Mint, Spink & Son, and Garrard & Co.
Number of Appointments: 107 Canadians; 9 Newfoundlanders.
The Distinguished Service Order
Distinguished Service Order obverse.
Origins: The Distinguished Service Order was created by Queen Victoria in 1886 as a reward for commissioned officers for distinguished war service. In some ways the DSO was meant as an award junior to being appointed a Companion of the Order of the Bath. During the South African War (1899–1902), the First and Second World Wars, and the Korean War, the DSO was awarded to officers for gallantry below that necessary for the award of the Victoria Cross.
Distinguished Service Order George V reverse.
Criteria: Awarded to members of the Royal Canadian Navy, Canadian Army, Royal Canadian Air Force, and Merchant Navy for gallantry or leadership under fire, defined as "meritorious or distinguished service in war."[3] The order was usually only awarded to majors and lieutenant-colonels, though appointments to more junior commissioned ranks were made — generally only in circumstances of significant acts of gallantry. From 1914 to 1916 the order was also awarded for services rendered under circumstances not considered to be "under fire."
Insignia: A gold cross 42 mm wide with curved tips and white enamelled fields, the obverse bearing a central roundel displaying either a Tudor or St. Edward's Crown in gold on a field of red enamel circumscribed by gold and green laurelled leaves. The reverse being identical to the obverse with the crown replaced by the Royal Cypher:
* • Victoria (1886–1901): VRI.
* • Edward VII (1901–10): EVIIR.
* • George V (1910–36): GVR.
* • George VI 1st type (1936–48): GRI.
* • George VI 2nd type (1948–52): GVIR.
* • Elizabeth II (1952 to present): EIIR.
Distinguished Service Order George VI 1st type reverse.
Suspender: A gold rectangular laurelled bar 31 mm wide with a raised border. An identical bar is mounted to the top of the ribbon as a brooch.
Ribbon: A crimson ribbon 29 mm wide edged with 3 mm of garter blue.
Bar: A gold bar 32 mm wide with a graduated line bisecting the length, displaying a Tudor Crown in the centre.
Postnominals: DSO.
Other: From 1886 to 1889 the insignia was made of carat gold while subsequent issues have been in silver gilt. All of the insignia awarded to Canadians were manufactured by Garrard & Co. Beginning in 1938, all awards of the DSO had the year of the award engraved on the reverse of the lower suspension.
Distinguished Service Order appointment warrant.
Number of Canadian Appointments: 1,162, 104 1st bars, 18 2nd bars.
* • South African War: 20.
* • First World War: 702 (88 1st bars, 15 2nd bars).
* • Second World War: 432 (17 1st bars, 2 2nd bars).
* • Korean War: 8 (6 GVIR 2nd types, 2 EIIR, and 1 1st bar).
The Imperial Service Order
Established by King Edward VII on August 8, 1902, the Imperial Service Order (ISO) was created to recognize long-serving senior civil servants (manager level and above) in both the United Kingdom and throughout the Commonwealth and British Empire. Senior civil servants in Britain were usually recognized with an appointment to the civil division of the Order of the Bath, and in the colonies it was the Order of St. Michael and St. George that was used to recognize colonial civil servants and British civil servants serving overseas.
Despite the pre-existence of two orders specifically intended for the civil service, their total membership was capped to a limited number of living members. This meant that people could only be appointed to the Order of the Bath or the Order of St. Michael and St. George when a vacancy occurred. With the rapid expansion of the British Empire in the last part of the nineteenth century, it became clear that another mechanism had to be developed to recognize those members of the civil service who had rendered long and meritorious service in the United Kingdom and especially in the less hospitable colonies. The solution to this was found in the creation of the Imperial Service Order. Along with the ISO, another award, the Imperial Service Medal, was established to recognize the service of junior civil servants.
Imperial Service Order Edward VII obverse.
The order was awarded to Canadians from its inception in 1902 until 1918, then again from 1933 to 1935, and finally from 1944 to 1946. In Canada it was typically awarded to the most senior civil servants at the level of deputy minister and to the heads of certain government agencies. One Newfoundlander was appointed to the order.
By the time the last Canadian was appointed to the order in 1946, the total membership throughout the Commonwealth and British Empire was limited to seven hundred. The British ceased awarding the ISO in 1995, though it is still used in certain Commonwealth countries.
Criteria: Twenty-five years of service in the British, dominion, or colonial civil service. It was not an automatic award, and nominations in Canada were put together by the clerk of the Privy Council and the undersecretary of state and submitted to the prime minister for transmission to the King. In "unhealthy" places the length of service required was a minimum of sixteen years (Canada was not classified as an unhealthy place).
Insignia: A six-pointed silver star 56 mm wide, in the centre of which is a gold disc bearing the Royal Cypher (EVIIR, GVR, or GRI), circumscribed with FOR FAITHFUL SERVICE in blue enamel and surmounted by an enamelled Imperial Crown. Queen Elizabeth II issues do exist, though none were awarded to Canadians.
Imperial Service Order 1st type appointment warrant.
Suspender: A small eyelet is affixed to the reverse of the crown and through this passes a suspender ring and the ribbon.
Ribbon: A watered ribbon 38 mm wide made up of equal parts red, blue, and red.
Postnominals: ISO.
Other: The ISO was manufactured by Elkington and Company of London and is found with sterling hallmarks on the reverse of the silver star of the order. The centre was made of 14-carat gold and occasionally bears a small stamp "14 K." Although issued unnamed, all ISOs were awarded with an appointment scroll embossed with the seal of the order and the name of the recipient.
Group of Harold Payne, deputy minister of national war services, 1943–48, containing a 1st type George VI Imperial Service Order.
Number of Canadian Appointments: 9.
* • Edward VII: 44.
* • George V: 20.
* • George VI (Type I): 33.
The Canada Medal
Canada Medal English obverse.
During the Second World War, a special government committee was established to administer honours in Canada. The Awards Coordination Committee (ACC) consisted of senior civil servants.[4] To remedy the lack of honours for Canadians, the ACC attempted on multiple occasions to establish a Canadian Order, only to be blocked by Prime Minister William Lyon Mackenzie King. Ultimately, the ACC was able to develop the Canada Medal as a universal Canadian honour to be awarded for a myriad of services. The genius behind the Canada Medal originated with Major-General Harry Letson.
The proposal set out such things as who was to receive the medal and what it would look like, among other important provisions:[5]
> 6. Eligibility: Personnel eligible to receive the medal shall be: (a) Citizens of Canada, whether civilians or members of the armed forces or of the Merchant Navy. (b) Citizens of other countries who have rendered valuable and meritorious services of the nature set forth in the next succeeding paragraph.
>
>
>
>
> 7 . Service Required: The medal may be awarded to persons named in the preceding paragraph for specially valuable and meritorious service of a high standard, faithful or zealous performance of ordinary duty not being sufficient in itself. There must be either:
>
> (a) special services of a high degree of merit, such as discharge of special duties superior to the person's ordinary work, or
>
> (b) highly meritorious performance of ordinary duties where these have entailed work of a specially trying character, or,
>
> (c) display of a high degree of initiative and forethought...[6]
>
>
>
After some questions from the King about the number of awards — he was greatly concerned that the Canada Medal would be too widely bestowed, thereby depriving it of any importance — the King approved the establishment of the honour. The entire Canada Medal project was formalized through Order-in-Council 7964 on October 14, 1943. Canadians were informed of the Canada Medal through a press release issued on October 17.
Canada Medal French obverse.
To deal with the King's concerns vis-à-vis the number of Canada Medals, it was agreed that:
>
>
>
> i) Quotas, to be on a half-yearly basis of one per two thousand instead of one per thousand. Based on present strength, this will limit each half-yearly list of Service awards to approximately 375 instead of 750;
>
>
>
>
> ii) publish CM list at the same time as Birthday Honours and New Year's Honours so that it will "not then be considered a 'consolation' award;"
>
>
>
>
> iii) first list to be issued on 1 January 1944;
>
>
>
>
> iv) no more than 50 percent of the total allotment to be awarded to officers."[7]
Public reaction was muted, war news being more important than recent institutional developments in Canada. Some papers were concerned about the "new" nature of the medal, postulating that if too many were awarded, it would become quite worthless. Toronto's Globe struck a balanced note, stating that the new medal would "win the approval both of keen Imperialists and ardent Canadian nationalists"[8] and stressing the "democratic flavor" of the whole project.
The Canada Medal was seen as a way to blend both the imperial honours system — particularly the orders of chivalry — with a distinctive Canadian award. The French-Canadian press was even more enthusiastic. La Presse especially endorsed the creation of a Canadian award, but like the Globe expressed concern that it might be awarded too generously.
The ACC originally planned for the first awards of the Canada Medal to be made on November 11, 1943. The intent was to recognize several outstanding recipients, and there is some evidence to suggest that they were to include His Majesty the King and Queen Elizabeth, Winston Churchill, Franklin Roosevelt, and the governor general, the Earl of Athlone, with the first full list to be published on February 1, 1944. However, this was not practical, and Cabinet was not prepared to proceed so quickly.[9]
At the November 3 Cabinet meeting it was decided that the Canada Medal would be awarded for outstanding service and not to those already in the highest positions. When Arnold Heeney, clerk of the Privy Council from 1940 to 1949 and Canada's most senior public servant during the Second World War, suggested that the ACC had already prepared to proceed with the awards, Mackenzie King recounted that
Canada Medal reverse.
> this got under my skin at once and I said it was the strongest of reasons why we should not proceed any further; that I objected altogether to these matters being forced on the Cabinet; that nothing should be done.... I]t would have been a scandal to permit this kind of thing to mar the face of the Government with the problems that are being confronting it at this time.... Perhaps the country itself will have little comment on the decision we have made. On the other hand, had we continued with the granting of decorations when there is no time to consider them, we would have a very bitter opposition aroused against us from coast to coast and in Parliament itself.[[10]
This was a distinctively inauspicious beginning for the Canada Medal.
Unlike 1943, there was no 1944 New Year's Honours List, or 1944 King's Birthday Honours List, and thus the newly minted Canada Medal had no outlet to be awarded. Finally, in December 1945, the prime minister announced that there would be a forthcoming Honours List on Dominion Day 1946. Again it was widely expected that the Canada Medal would be awarded, along with the various levels of the Order of the British Empire and the British Empire Medal. Prior to the ACC soliciting nominations for honours from various government departments, the military, and civilian organizations, a notice was sent to all deputy ministers and eventually to the press, stating: "The Government's intentions with respect to the award of the Canada Medal and the establishment of a distinctive Canadian Order will be announced in due course.[11]
Again the Canada Medal was quashed, and it seemed as though it would never be awarded. Prime Minister Mackenzie King recorded in his diary:
> I told the Cabinet I had never experienced more pain and anguish over any public matter than I have on anything that had to do with decorations and honours. I said that personally I was against them and upon conviction, for honours that were done one, multitudes were ignored who are more worthy.[12]
Canada Medal nomination form.
Without support from the prime minister, the entire project remained moribund. As soon as the last Canadian civilian honours list was issued in 1946, interest in the Canada Medal evaporated. It would be mentioned in the House of Commons on only a few more occasions. Neither Liberal nor Conservative governments were interested in the award.
With the creation of the Order of Canada on April 17, 1967, and the issuing of Order-in-Council 388, the Canada Medal ceased to be, never having been awarded to a single soul.
Origins: Intended to serve as a Canadian award of merit to replace all the imperial orders of chivalry, which had traditionally been awarded to Canadians for civil and military service.
Criteria: Awarded for "specially valuable and meritorious service of a high standard, faithful or zealous performance of ordinary duty not being sufficient itself. There must be either a) special services of a high degree of merit, such as discharge of special duties superior to the person's ordinary work, or b) highly meritorious performance of ordinary duties where these have entailed work of a specially trying character, or c) display of a high degree of initiative and forethought."
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing an effigy of King George VI crowned, circumscribed by the text GEORGIVS VI D:G: BR: OMN: REX ET INDIAE IMP:. The reverse bearing the shield of arms of Canada surmounted by a Tudor Crown and surrounded by maple leaves, with a small scroll at the bottom bearing CANADA. The medal was designed by Thomas Shingles.
Suspender: An inverted fleur-de-lys holding a plaque bearing the word MERIT or MERITE in relief.
Ribbon: A 32 mm wide ribbon made up of equal portions of red, white, and red.
Bars: None.
Postnominals: CM or MduC.
Other: All medals were stamped with SPECIMEN on the rim.
Number of Awards: 7 English and 7 French specimens were struck by the Royal Canadian Mint.
# 5
IMPERIAL GALLANTRY DECORATIONS
For much of Canada's post-Confederation history, the imperial gallantry decorations served as the primary mechanism for recognition of gallant or brave conduct. Their use tended to be restricted to wartime, namely, the South Africa War (1899–1901), the First and Second World Wars, the Korean War, and to a very limited extent, U.N. peacekeeping missions. Unlike other honours, many gallantry decorations and awards were accompanied by an annuity, the amount and criteria for which changed over the history of the decoration.
The development of imperial gallantry decorations and awards was sporadic and occurred over more than a century. Most honours were not created as part of an overall system; rather, they were established to fill an immediate gap and to ensure that brave service in a particular field was recognized. This is in contrast to the Canadian honours system in which developments have tended to come in tranches, beginning with the creation of the Decorations for Bravery in 1972, followed by the Military Valour Decorations in 1993. The Canadian honours system has benefited immeasurably from the experiences of our Commonwealth cousins.
Looking at the system of imperial gallantry decorations and medals today, it seems quite odd that the honours a serviceman was eligible for depended largely on his rank and in what service he was in. This is in stark contrast to the Canadian honours system where the same set of awards is used for all ranks and all services. Indeed, after the First World War and the introduction of the service- and rank-specific gallantry awards, namely, the Distinguished Service Cross, Military Cross, Distinguished Flying Cross, and their medal equivalents the Distinguished Service Medal, Military Medal, and Distinguished Flying Medal, it was the wish of King George V that these awards, along with the Distinguished Conduct Medal and Conspicuous Gallantry Medal, be abolished and replaced with a pair of new honours. This had been recommended by the Report of the Committee on Honours and Rewards in War, which in 1923 had proposed the creation of a Distinguished Service Decoration and a Distinguished Service Medal. Despite the King's strong support and that of the committee, the three services jealously resisted any changes.
Throughout the twentieth century there was an increasing requirement to recognize civilians who acted gallantly. It was no longer just members of the military who were eligible for official recognition. A variety of organizations in Britain such as the Royal Humane Society, Lloyd's of London, and the Society for the Protection of Life from Fire, sought to recognize civilian bravery; however, there were no specific honours of the Crown bestowed to recognize similar acts until the Albert Medal was created in 1866. This was followed by the Edward Medal, the King's Police Medal, the Empire Gallantry Medal, and then the George Cross and the George Medal. The Edward Medal is of particular interest since it was intended for labourers and those working in difficult circumstances at the bottom end of the social scale. The use of the Order of the British Empire and British Empire Medal to recognize gallantry is also worthy of consideration and has been examined in chapter 4.
In Canada the George Medal, Albert Medal, Air Force Medal, Edward Medals, King's Police Medal, King's Commendation for Brave Conduct, and King's Commendation for Valuable Service in the Air were awarded sporadically during peacetime, though such awards were rare. The gallantry decorations and medals bestowed upon Canadians prior to 1967 differ from the modern Military Valour Decorations and decorations for bravery in that the branch of the military one served in — RCN, Army, or RCAF — and whether or not you were an officer determined the honour for which you were eligible. This was true for all of the gallantry awards, save the Victoria Cross, Albert Medal, George Cross, and George Medal, which were totally rank- and service-"blind" and could be earned by anyone. The Royal Red Cross was unique in that it was only bestowed upon women.
After the end of the Second World War, the restrictive nature of Canadian honours policy almost precluded any civilian recognition, while those in the military remained eligible for gallantry awards — but not those for meritorious service such as the Order of the Bath or Order of the British Empire. In 1956 a more liberal honours policy was introduced via Cabinet Directive No. 30 (see chapter 3). This allowed for civilians to be recognized for their gallant acts with the George Cross, the George Medal, the Queen's Commendation for Brave Conduct, and the Order of the British Empire with gallantry emblem. Members of the Canadian Armed Forces remained eligible for the host of service- and rank-specific gallantry decorations.
All except one of the decorations and awards covered in this chapter have been supplanted by a variety of Canadian honours in the post-1967 period, the exception being the Mention in Dispatches, though it was revived in a Canadian form in 1990. The Canadian Victoria Cross, Cross of Valour, Military Valour Decorations, Decorations for Bravery, and Meritorious Service Decorations, along with the Order of Military Merit and Order of Merit of the Police Forces, overlap in criteria and purpose with those honours examined in this chapter. In 1993 the British reformed their honours system to be rank-blind, in that those who, on account of not being officers, were once only eligible for the DCM, CGM, DSM, MM, DFM, and AFM are now eligible for the equivalent that were once only conferred on officers.
The number of each honour bestowed has been taken from the important work done by Surgeon-Commander J.F. Blatherwick, CM, OBC, CD, and Hugh Halliday, who have devoted years to compiling the names and details of so many brave Canadians.
The Victoria Cross
Origins: Queen Victoria created the Victoria Cross following the end of the Crimean War in 1856 to recognize the most outstanding acts of gallantry. The impetus behind the establishment of the VC seems to be, in part, connected to discussions in Parliament and in the press that called for a reward to recognize soldiers and officers who had rendered daring, brave, and gallant acts throughout the two-year war with Russia. While officers were eligible for appointment to the Order of the Bath, men for the award of the Distinguished Conduct Medal and sailors for the award of the Conspicuous Gallantry Medal, established in 1854 and 1855 respectively, there was little in the way of an honours system that sought to recognize a diverse array of military and civil acts of the highest order.
Only one VC was awarded for actions on Canadian soil, that to Private Timothy O'Hea of the Rifle Brigade. O'Hea extinguished a fire in an ammunition train during the Fenian Raid of 1866 in Danville, Quebec. This award was highly unusual since O'Hea's gallantry was not rendered in the face of the enemy.
Victoria Cross obverse.
Only five other VCs were awarded under similar circumstances of "extreme danger, such as the occurrence of a fire on board ship... or under any other circumstances in which, through courage and devotion displayed, life or public property might be saved."[1] Since 1881 the VC can only be awarded for gallantry in the face of an enemy.
Canadians received a total of 94 of the 1,357 Victoria Crosses awarded between its inception and 2014.[2] There are no Canadian recipients still living. The first Canadian to receive it was Alexander Dunn, a native of Upper Canada, who displayed outstanding valour during the famous Charge of the Light Brigade at Balaclava in the Crimean War. The second VC to be awarded to a Canadian was also the first conferred on a person of colour: William Hall of Nova Scotia received it for service in the Indian Mutiny while serving in the Royal Navy. Perhaps the most famous Canadian VC recipient is Billy Bishop, the noted Great War flying ace and later air marshal in the Royal Canadian Air Force. The VC was replaced with the Canadian VC (see chapter 18).
Criteria: For most conspicuous bravery or some daring or pre-eminent act of valour or self-sacrifice or extreme devotion to duty in the presence of the enemy.
Insignia: A bronze straight-armed cross pattée with raised edges, measuring 38 mm across. The obverse depicts a lion guardant standing upon the Royal Crown, and below the crown a scroll bears the inscription FOR VALOUR. The reverse is plain with raised edges. The insignia was most likely designed by H.H. Armstead of Hancocks.
Suspender: A small semicircle with a hole in the centre is located on the uppermost arm of the cross. Through this passes a small bronze link connected to a V below a straight bar ornamented with laurel leaves. On the reverse of the bar is engraved the rank, name, and unit of the recipient.
Victoria Cross reverse.
Ribbon: A crimson ribbon, 38 mm wide. A recipient wears a single miniature of the VC on his undress ribbon. A second award is denoted by a second miniature VC worn on the undress ribbon. A blue ribbon was used for naval awards until 1918.
Bars: Additional awards of the VC are signified by a bronze bar ornamented with laurel leaves.
Naming: Hand-engraved capitals on the reverse of the suspender with the recipient's name and rank and date of action in the centre of the reverse.
Postnominals: VC.
Other: Hancocks & Co produced numbered replicas made from commercial bronze.
Number of Awards to Canadians: 94
* • 4 pre–South African War.
* • 70 First World War.
* • 16 Second World War.
The George Cross
Origins: At the beginning of the Second World War there was a variety of honours that recognized gallant or brave acts rendered in situations not necessarily in the face of the enemy. The Empire Gallantry Medal, Albert Medal, and Edward Medal all existed to recognize courageous acts, primarily in non-combat settings. The intensity of the air war being waged on the United Kingdom, and the greatly enhanced role of civilians and those working on the home front, necessitated the creation of a decoration on the same level as the Victoria Cross,
George Cross obverse.
George Cross reverse.
though for non-combat situations, to recognize both civilians and those serving in the mili-tary who rendered outstanding acts of gallantry. The George Cross and George Medal were both created by King George VI in September 1940; indeed, the King took a personal role in the design of the insignia and development of the criteria pertaining to the awards. With the creation of the GC, the Empire Gallantry Medal was cancelled and all living holders of the EGM were permitted to exchange their insignia for a GC. In 1971 all living recipients of the Albert Medal and Edward Medal were invited to exchange their insignia for a GC.
Criteria: To recognize acts of the greatest heroism or of the most conspicuous courage in circumstances of extreme danger.
Insignia: A silver Geneva cross 45 mm wide and tall, the obverse depicting in the centre St. George slaying the dragon, circumscribed by the legend FOR GALLANTRY and a small rose at the base. The angle of each arm of the cross contains the Royal Cypher of King George VI, GRI. The reverse is plain. The GC was designed by George VI and F.W. Barry.
Suspender: A slotted bar 39 mm wide displaying laurel leaves. A small eyelet is attached to the bottom of the bar through which passes an oblong ring that attaches to another eyelet on the top of the cross.
Ribbon: A garter blue ribbon 38 mm wide. A small silver GC is worn in the centre of the undress ribbon.
Bars: None to Canadians.
Group of medals to Colonel A.R.C. Butson, GC, OMM, CD.
Naming: Engraved capitals, number, and abbreviated rank (if any) along with full given name, surname, and date the award was announced in the London Gazette, all hand-engraved in capitals.
Postnominals: GC.
Other: Replaced by the Cross of Valour.
Number of Awards to Canadians: 10.2 were exchanged from Albert Medals while 1 was converted from an Empire Gallantry Medal).
The Albert Medal
Albert Medal for Gallantry at Sea in gold obverse.
Origins: Established by Queen Victoria in 1866 as a single-level honour, the Albert Medal was broadened by 1867 into a two-class decoration to recognize acts of gallantry at sea. In 1877 the Albert Medal was expanded in scope to include acts of gallantry on land. From its inception this medal was open to both civilians and members of the military and was the first official honour bestowed by the Crown intended to recognize brave acts rendered by civilians. As noted by P.E. Abbott and J.M.A. Tamplin, "the standard of gallantry qualifying for an award has always been very high and it seems that the criterion adopted has been that the recipient's risk of death had to be greater than chances of survival, and that in the case of the gold medal 1st class] the risk had to be altogether exceptional."[[3]
Albert Medal for Gallantry at Sea in bronze obverse.
There being a significant degree of overlap between the criteria of this award and the George Cross and George Medal, awards of the Albert Medal in Gold ceased in November 1949. Awards of the Albert Medal (formerly known as the AM 2nd class) ceased in 1971. Although initially divided into two classes, 1st and 2nd, King George V approved a change in 1917 that designated the Albert Medal 1st class the "Albert Medal in Gold," while the Albert Medal 2nd class became simply the "Albert Medal."
Criteria: The criteria can be divided into two parts, dependent on whether or not the act of bravery was rendered at sea or on land:
* • Gallantry at Sea: To recognize daring and heroic actions performed by mariners and others to prevent the loss of life and to save the lives of those who are in danger of perishing by reason of wrecks and perils of the sea; the 1st class being confined to cases of extreme and heroic daring, while the 2nd class being limited to other acts of heroic daring.
* • Gallantry on Land: to recognize daring and heroic actions performed on land to prevent the loss of life and to save the lives of those who are in danger of perishing; the 1st class being confined to cases of extreme and heroic daring, while the 2nd class being limited to other acts of heroic daring.
Albert Medal for Gallantry at Sea in bronze reverse.
Insignia (for Gallantry at Sea): An oval-shaped gold (1st class/in gold) or bronze (2nd class/Albert Medal) insignia 32 mm wide and 38 mm high, bearing in the centre a monogram of the letters V and A interlaced (for Victoria and Albert) with an upright anchor, the background enamelled in translucent blue with a starburst pattern beneath, the entire device encircled by a garter belt bearing the text FOR GALLANTRY IN SAVING LIFE AT SEA. The insignia is topped with the crown worn by Prince Albert. The reverse is plain other than naming details. The reverse bears the text PRESENTED IN THE NAME OF HIS/HER MAJESTY or AWARDED BY HIS/HER MAJESTY along with details of the recipient and details of the gallant action.
Insignia (for Gallantry on Land): An oval-shaped gold (1st class/in gold) or bronze (2nd class) Albert Medal insignia 32 mm wide and 38 mm high, bearing in the centre a monogram of the letters V and A, the background enamelled in translucent red with a starburst pattern beneath, the entire device encircled by a garter belt bearing the text FOR GALLANTRY IN SAVING LIFE ON LAND. The reverse is the same as that of the Albert Medal for saving life at sea.
Suspender: A ring 11 mm in diameter passes through the orb of the crown through which the ribbon passes. The ring is gold for the 1st class and bronze for the 2nd class.
Ribbon: Four different ribbons were issued, and they were dependent on the class of award and division (sea or land):
* • Albert Medal 1st Class (Sea): Watered white, 35 mm wide with 5 garter blue stripes 3 mm wide.
* • Albert Medal 1st Class (Land): Watered white, 35 mm wide with 5 red stripes 3 mm wide.
* • Albert Medal 2nd Class (Sea): Watered white, 35 mm wide with a central garter blue stripe 8 mm wide.
* • Albert Medal 2nd Class (Land): Watered white, 35 mm wide with a central red stripe 8 mm wide.
Albert Medal for Gallantry on Land in bronze obverse.
Bars: Provision was made for the bestowal of a bar; however, none were ever awarded.
Naming: Engraved in upper and lower case, bearing the full given and surname of the recipient along with details of the action.
Postnominals: AM.
Albert Medal for Gallantry on Land in bronze reverse.
Other: Presented in a rectangular maroon leatherette case embossed with the Royal Cypher of the reigning sovereign on the lid.
Number of Awards to Canadians: 22 (14 2nd class for gallantry at sea, 1 1st class for gallantry on land, 7 2nd class for gallantry on land). In addition to this, 2 1st-class and 2 2nd-class Albert Medals for gallantry at sea were awarded for the Halifax Explosion to members of the Royal Navy.
The Empire Gallantry Medal
Empire Gallantry Medal obverse.
Origins: The Empire Gallantry Medal was created by King George V in 1922 to replace the Medal of the Order of the British Empire. While originally styled "The Medal of the Order of the British Empire (for Gallantry)," this was superseded by the more elegant sounding Empire Gallantry Medal. This medal was cancelled in 1940 and was superseded by the George Cross. In 1971 living recipients of the EGM were permitted to exchange their awards for a George Cross. Only one person associated with Canada was awarded the EGM, Aircraftsman 1st Class Ernest Frost, who served in the Royal Air Force. In total only 130 were awarded for the entire British Empire.
Criteria: Awarded for acts of gallantry.
Insignia: A circular silver medal 36 mm in diameter, the obverse depicting a seated effigy of Britannia holding a trident and circumscribed by the text FOR GOD AND THE EMPIRE with GALLANTRY at the base. The reverse displays the Royal Cypher surmounted by a crown in the centre (GRIV) surrounded by four lions. The reverse of the medal was changed following the death of King George V and subsequently displayed the Royal Cypher (GRI) surmounted by a crown in the centre surrounded by two lions on each side with the text INSTITUTED BY KING GEORGE V. Recipients of the EGM wore a silver laurel branch on the ribbon and undress ribbon of the medal to signify that the award was for gallantry.
Ribbon: Same as for the British Empire Medal.
Suspender: A claw suspender composed of a trio of stylized oak leaves attached to a straight suspension bar.
Bars: The EGM could be awarded with a bar; none to Canadians.
Naming: Engraved capitals, full given name and surname.
Postnominals: EGM.
Number of Awards to Canadians: 1.
The Distinguished Conduct Medal
Distinguished Conduct Medal trophy arms obverse.
Origins: The Distinguished Conduct Medal was created during the Crimean War to recognize gallant acts rendered by soldiers and non-commissioned officers serving in the army. The DCM was created to meet the need to grant "an official gallantry medal."[4] Queen Victoria established the DCM in 1854, and it endured a variety of changes over its nearly 150-year history. With the creation of the Victoria Cross in 1856, the DCM became the second-highest gallantry award open to those of warrant officer rank and below. A special dominion and colonial issue of the DCM was created in 1894; however, none were awarded under the regulations of the 1894 Royal Warrant, though a number of specimens are known to exist. The modern Canadian equivalent is the Star of Military Valour.
Criteria: Awarded to warrant officers, non-commissioned officers and men in recognition of distinguished conduct in the field.
Insignia: A circular silver medal 36 mm in diameter; there have been six obverse types awarded to Canadians:
* • Edward VII type: Edward VII uncrowned in a field marshal's uniform circumscribed by the legend EDWARDVS VII REX IMPERATOR.
* • George V 1st type: George V uncrowned in a field marshal's uniform circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND:IMP:.
* • George V 2nd type: George V crowned in coronation robes circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND:IMP:.
* • George VI 1st type: George VI crowned circumscribed by the legend GEORGIVS VI D:G: BR: OMN: REX ET INDIAE IMP:.
* • George VI 2nd type: George VI crowned circumscribed by the legend GEORGIVS VI DEI GRA: BRITT. OMN: REX FID: DEF+.
Distinguished Conduct Medal Edward VII obverse.
There were two reverse types of this medal awarded to Canadians:
* • 1st type: plain with CANADA arched above the text FOR DISTINGUISHED CONDUCT IN THE FIELD displayed on 4 lines with a decorative broken line at the base.
* • 2nd type: plain with the text FOR DISTINGUISHED CONDUCT IN THE FIELD displayed on 4 lines with a decorative broken line at the base.
Suspender: A swivelling claw-footed suspender with an ornate scroll bar.
Ribbon: A crimson ribbon 32 mm in width divided in the centre by a 10 mm stripe of dark blue.
Bars: A silver bar 33 mm wide with raised edge displaying laurel leaves in the centre slides on the ribbon.
Distinguished Conduct Medal George VI 1st type obverse.
Distinguished Conduct Medal reverse.
Naming: Impressed capitals, number, abbreviated rank, initials, surname, and unit.
Postnominals: DCM.
Other: One Canadian issue of the "trophy of arms" DCM is known to have sold at public auction; however, further details of this medal are not known. The CANADA reverse is only known to have been issued as specimens. The George V 2nd type was issued to a very small number of Canadians who earned the DCM during the First World War and had their medals replaced (due to loss or theft) during the 1930s. No Elizabeth II issues of this medal were awarded to Canadians. In 1993 the British replaced this decoration with the Conspicuous Gallantry Cross.
Number of Awards to Canadians: 2,132, 38 1st bars and 1 2nd bar.
* • South Africa: 16.
* • First World War: 1,947 (36 1st bars; 1 2nd bar).
* • Second World War: 162 (1 1st bar).
* • Korean War: 7 (1 1st bar).
The Conspicuous Gallantry Medals
Conspicuous Gallantry Medal George V obverse.
Origins: The Conspicuous Gallantry Medal was established by Queen Victoria initially in 1855 as a modified version of the Meritorious Service Medal with a special reverse displaying the text FOR CONSPICUOUS GALLANTRY. Initially, this medal was awarded only to ratings and petty officers of the Royal Navy and the equivalent ranks in the Royal Marines for conspicuous gallantry during the Crimean War but was discontinued after the establishment of the Victoria Cross in 1856. The decoration was revived in 1874 toward the end of the Ashanti War. Being one grade below the Victoria Cross, very few were awarded.
A Conspicuous Gallantry Medal (Air) was established on January 15, 1943, to fill the gap of recognition that existed for non-commissioned officers and airmen between the Distinguished Flying Medal and the Victoria Cross. While the navy had the Conspicuous Gallantry Medal (Sea) and the army the Distinguished Conduct Medal, there was no 2nd-level gallantry award for those serving in the various Commonwealth air forces. It is no coincidence that this decoration was created during one of the most intense periods of air operations in the history of modern warfare. The Star of Military Valour has replaced the CGM (see chapter 18).
Conspicuous Gallantry Medal for Air George VI obverse.
Conspicuous Gallantry Medal reverse.
Criteria: Dependent on branch of service. For the navy it was awarded to recognize seamen and non-commissioned officers who distinguished themselves by acts of conspicuous gallantry in action with the enemy. For the air force it was awarded to recognize airmen and non-commissioned officers who distinguished themselves by acts of conspicuous gallantry while flying in active operations against the enemy.
Insignia: A circular silver medal 36 mm in diameter, the obverse carrying a crowned effigy of King George VI circumscribed by the legend GEORGIVS VI D:G: BR: OMN: REX ET INDIAE IMP:. The reverse displays a Victorian-style Royal Crown at the top and the text FOR CONSPICUIOUS GALLANTRY on three lines with a small decorative broken line at the bottom, the entire text and crown encompassed between two branches of a laurel wreath.
Suspender: A swivelling claw-footed suspender with a straight bar.
Ribbon: Two separate ribbons were issued with this medal, the colour of ribbon being dependent on whether the medal was bestowed for action at sea or in the air.
* • Sea: A white ribbon 32 mm wide edged with 3 mm of dark blue on each side.
* • Air: A pale blue ribbon 32 mm wide edged with 3 mm of dark blue on each side.
Bars: None to Canadians.
Naming: Engraved or impressed capitals, number, abbreviated rank, initials, surname, and RCN or RCAF.
Postnominals: CGM.
Other: In 1993 the British replaced this decoration with the Conspicuous Gallantry Cross.
Number of Awards to Canadians: 21 (2 sea and 19 flying). All Second World War Awards.
The Royal Red Cross
Royal Red Cross George V obverse.
Origins: Created by Queen Victoria in April 1883 to reward nurses serving in military theatres or settings, the decoration was expanded to include a junior award, known as the Associate Royal Red Cross, in 1915. In 1917 provisions were made for the bestowal of a bar to the Royal Red Cross (holders of the Associate Royal Red Cross are promoted to the RRC in place of receiving a bar). The 1st class is formally known as a Member of the Royal Red Cross, while the 2nd class is formally known as an Associate of the Royal Red Cross.
Associate Royal Red Cross George V obverse.
Throughout the late Victorian period there was a gradual professionalization of the military medical services, which had largely been attached to naval and military organizations in an ad hoc manner up to this point. The pre-eminent role of women in providing nursing services — one of the most notable being Florence Nightingale during the Crimean War — required that an appropriate reward be developed to recognize the special services and skills rendered by women in medical care provision roles. Men were, of course, eligible for a host of gallantry decorations and appointment to various orders in recognition of their service; however, women were largely omitted from the honours system, a trend that began to change with the establishment of the Royal Red Cross, and later the Order of the British Empire, which was from the outset bestowed upon both sexes.
Associate Royal Red Cross reverse.
Criteria: Awarded to those who have shown exceptional devotion and competency in the performance of nursing duties over a continuous and long period of time, or who have performed some very exceptional acts of bravery and devotion at their posts of duty. The Royal Red Cross (Member) was awarded to fully trained nurses of an officially recognized nursing service, with the overall membership limited to a maximum of 2 percent of the total nurses serving, while the Associate of the Royal Red Cross (Associate) in addition to being open to fully trained nurses was also open to assistant nurses, nurse probationers, Voluntary Aid Detachment personnel, and those serving in a recognized nursing service. The overall membership of the ARRC was limited to a maximum of 5 percent of the total nurses serving. Until 1976 both awards were limited to women only. Appointments to the Order of Military Merit or the Meritorious Service Decorations are the equivalent modern awards.
Insignia: Member of the Royal Red Cross: a gold cross pattée 35 mm in width, the obverse enamelled in translucent red bearing in the centre the effigy of the reigning Sovereign. The reverse carries on each arm one of the words FAITH, HOPE, CHARITY, with the date 1883 on the lowermost arm. The reverse is plain other than the Royal Cypher of the reigning Sovereign, which is displayed in the centre.
Associate of the Royal Red Cross: a frosted silver cross pattée 35 mm in width, the obverse enamelled in translucent red bearing in the centre the effigy of the reigning Sovereign. The reverse carries on each arm one of the words FAITH, HOPE, CHARITY, with the date 1883 on the lowermost arm. The centre of the reverse carries the Royal Cypher of the reigning Sovereign.
There have been three obverse types awarded to Canadians (both RRC and ARRC):
* • Victoria veiled effigy.
* • George V uncrowned effigy.
* • George VI uncrowned effigy.
There have been four reverse types awarded to Canadians (both RRC and ARRC):
* • Victoria: VRI.
* • George V: GRI.
* • George VI 1st type: GRI.
* • George VI 2nd type: GVIR.
Suspender: A ring is attached to an eyelet mounted on top of the insignia. Through this ring passes a sewn piece of ribbon that is affixed to the bow.
Ribbon: A garter blue ribbon 26 mm side edged on each wide with 6 mm of red.
Bars: A gold bar with red enamel.
Naming: None. Insignia awarded after 1939 are engraved with the year of the award on the reverse.
Postnominals: RRC or ARRC.
Other: All insignia awarded to Canadians were manufactured by Garrard & Co.
Number of Awards to Canadians: 124 RRC and 4 1st bars, 410 ARRC.
* • Royal Red Cross
* ɣ South Africa: 1.
* ɣ First World War: 66, 4 1st bars.
* ɣ Second World War: 56.
* ɣ Korean War: 1.
* • Associate Royal Red Cross
* ɣ First World War: 275, 17 elevations.
* ɣ Second World War: 134, 3 elevations.
* ɣ Korean War: 1, 1 elevation to RRC.
The Distinguished Service Cross
Distinguished Service Cross George VI 1st type obverse.
Distinguished Service Cross reverse.
Origins: Originally styled "The Conspicuous Service Cross," this decoration was created in 1901 to recognize subordinate officers and warrant officers serving in the naval forces for distinguished services in the face of the enemy. The criteria were changed in 1914 to make all officers holding the rank of lieutenant-commander and below eligible for the decoration. A bar was instituted in 1916. Until 1918 the DSC was also used to recognize those serving in the Royal Naval Air Service. The modern Canadian equivalent is the Medal of Military Valour.
Criteria: To recognize those who render meritorious or distinguished services in the face of the enemy.
Insignia: A matte silver cross pattée 40 mm in width bearing the Royal Cypher in the centre circumscribed by a raised circle. There have been three obverse types awarded to Canadians:
* • George V issue: GRI in script.
* • George VI 1st type: GRI in block letters.
* • George VI 2nd type: GVIR in script.
* • Elizabeth II issue: EIIR in block letters.
The reverse is plain, occasionally found with British sterling hallmarks impressed on the lower arm.
Suspender: An eyelet is mounted on top of the cross through which passes a heavy gauge ring 19 mm in diameter through which the ribbon passes.
Ribbon: A navy blue ribbon 35 mm wide with a central white stripe 11 mm wide.
Bars: A plain matte silver bar with ends flared in the same fashion as the DSC insignia, with a Tudor Crown in the centre.
Naming: None, though privately engraved examples do exist. Insignia and bars presented after 1940 are dated with the year of the award on the lower part of the reverse.
Postnominals: DSC.
Other: Insignia awarded to Canadians were all made by Garrard & Co. or Spink & Son.
Number of Awards to Canadians: 131 (14 1st bars, 3 2nd bars).
* • First World War: 1.
* • Second World War: 121 (17 1st bars, 3 2nd bars).
* • Korean War: 9 (1 1st bar).
The Military Cross
Military Cross George VI 1st type obverse.
Origins: The Military Cross was established by King George V in December 1914 to recognize distinguished and meritorious services rendered by officers holding the rank of captain or below and warrant officers. Until 1918 the MC was also used to recognize those serving in the Royal Flying Corps.
Criteria: For distinguished and meritorious service. In 1920 the Royal Warrant was changed to specify that the MC could only be given for gallant and distinguished services in action. This was in part in response to the fact that during the First World War a number of MCs were bestowed upon officers for services not necessarily rendered in the face of the enemy. The modern Canadian equivalent is the Medal of Military Valour.
Insignia: A plain silver cross 45 mm wide and tall bearing at the end of each arm a Tudor Crown, the centre of the cross displaying a smaller cross and the Royal Cypher. Four different obverse types have been awarded to Canadians:
* • George V issue: GRI (overlapping Royal Cypher).
* • George VI 1st type: GRI (letters side by each).
* • George VI 2nd type: GVIR.
* • Elizabeth II issue: EIIR.
The reverse of the insignia is plain. It is believed that the MC was designed by King George V.
Military Cross Elizabeth II obverse.
Military Cross reverse.
Suspender: A plain flat slotted suspender 37 mm wide bearing a small eyelet in the centre through which passes a ring that attaches to an eyelet mounted on the top of the insignia.
Ribbon: A watered white ribbon 32 mm wide with a central stripe of purple.
Bars: A silver bar 33 mm wide displaying in the centre a Tudor Crown. The bar slides over the ribbon.
Naming: None, though privately engraved examples do exist. Insignia and bars presented after 1938 are dated with the year of award on the lower part of the reverse.
Postnominals: MC.
Other: All insignia awarded to Canadians were struck by the Royal Mint.
Number of Awards to Canadians: 3,596 (307 1st bars, 17 2nd bars).
* • First World War: 2,885 (294 1st bars, 16 2nd bars).
* • Second World War: 678 (13 1st bars, 1 2nd bar).
* • Korean War: 33.
The Distinguished Flying Cross
Distinguished Flying Cross George VI 1st type obverse.
Origins: The Distinguished Flying Cross was established as part of a family of decorations to provide recognition for the newly created Royal Air Force, which came into being on April 1, 1918. Throughout the First World War members of the Royal Flying Corps and Royal Naval Air Service, the precursors to the RAF, were entitled to the various army and navy decorations respectively. With the air force created as a separate service, it was necessary to initiate a comparable set of decorations to recognize gallantry in an air setting. The DFC, DFM, AFC, and AFM were all created by King George V in June 1918. The Canadian Air Force, established in 1920 and renamed the Royal Canadian Air Force in 1924, was eligible for this group of awards until 1968. The modern Canadian equivalent is the Medal of Military Valour.
Criteria: Awarded to commissioned officers and warrant officers who perform acts of valour, courage, or devotion to duty while flying in active operations against the enemy.
Insignia: A silver cross flory terminated in the horizontal and base bars with bombs, the upper bar terminating with a rose, surmounted by another cross composed of airplane propellers charged in the centre with a rose superimposed with the letters RAF within a roundel, itself surrounded by a winged wreath of laurels ensigned by an Imperial Crown. The reverse displays an uncrowned Royal Cypher of the reigning Sovereign and the date 1918. The DFC was designed by E. Carter Preston. There have been four types of this decoration awarded to Canadians, the differences relating to the Royal Cypher displayed on the reverse:
* • George V issue: GRI with the arm of the R looping through the eye of the "8" in 1918.
* • George VI 1st type: GRI without the arm of the R intersecting with the "8" in 1918.
* • George VI 2nd type: GVIR.
* • Elizabeth II issue: EIIR.
Distinguished Flying Cross George VI 1st type reverse.
Suspender: A highly decorative suspender composed of two sprigs of laurel that graduate into a straight suspender. The suspender is attached to the top of the cross via a small oblong ring that passes through an eyelet mounted on top of the insignia.
Ribbon: A white ribbon 32 mm wide displaying purple stripes on a 45 degree angle, 3.175 mm apart. Until 1919 the stripes were horizontal.
Bars: A 33 mm wide silver bar with a flying eagle centred; the bar slips over the ribbon.
Naming: None, though privately engraved examples do exist. Insignia and bars presented after 1940 are dated with the year of the award on the lower part of the reverse.
Postnominals: DFC.
Other: Crosses awarded during the First World War were made by John Pinches, while those awarded to Canadians during the Second World War and the Korean War were made by the Royal Mint.
Number of Awards to Canadians: 4,019 (213 1st bars, 6 2nd bars).
* • Second World War: 4,018 (213 1st bars, 6 2nd bars).
* • Korean War: 1.
The Air Force Cross
Air Force Cross George VI 1st type obverse.
Origins: The Air Force Cross is part of the family of air force decorations created in June 1918 after the establishment of the Royal Air Force. The modern Canadian equivalent is the Meritorious Service Cross or Medal (military division).
Criteria: Awarded to commissioned officers and warrant officers who perform acts of valour, courage, or devotion to duty while flying, though not in active operations against the enemy.
Insignia: A silver cross in the form of a thunderbolt, the arms conjoined by wings, the base bar terminating with a bomb, surmounted by another cross composed of airplane propellers, the four ends inscribed with the letters GVRI (the elements of the Sovereign's Royal Cypher). In the centre of the roundel appears a representation of Hermes mounted on a hawk in flight bestowing a wreath. The reverse is plain, displaying in the centre an uncrowned Royal Cypher for the reigning Sovereign above the date 1918. There have been four types of this decoration awarded to Canadians:
* • George V issue: 1 letter of each part of the Royal Cypher GVRI on the arms of the cross. The reverse displaying in the centre GRI with the arm of the R looping through the eye of the "8" in 1918.
* • George VI 1st type: 1 letter of each part of the Royal Cypher GVIRI on the arms of the cross. The reverse displays in the centre GRI without the arm of the R intersecting with the "8" in 1918.
* • George VI 2nd type: 1 letter of each part of the Royal Cypher GVIR on the 3 uppermost arms of the cross. The reverse displays in the centre GVIR.
* • Elizabeth II issue: 1 letter of each part of the Royal Cypher EIIR on the uppermost arms of the cross. The reverse displays in the centre EIIR.
Air Force Cross George VI 1st type reverse.
Suspender: A highly decorative suspender composed to two sprigs of laurel that graduate into a straight suspender. The suspender is attached to the top of the cross via a small oblong ring that is integral to the main insignia (not a loose ring as for the DFC).
Ribbon: A white ribbon 32 mm wide displaying red stripes on a 45 degree angle, 3.175 mm apart. Until 1919 the stripes were horizontal.
Bars: A 33 mm wide silver bar with a flying eagle centred; the bar slips over the ribbon.
Naming: None, though privately engraved examples do exist. All insignia and bars awarded after 1939 are dated with the year of the award on the lower part of the reverse. Awards from 1939 are engraved with the year, while subsequent issues are engraved.
Postnominals: AFC.
Other: Crosses awarded during the First World War were made by John Pinches, while those awarded to Canadians during the Second World War and Korean War were made by the Royal Mint.
Number of Awards to Canadians: 454 (1 1st bar).
* • Second World War: 428 (1 1st bar).
* • Korean War: 4.
* • Postwar: 22.
The George Medal
George Medal reverse.
Origins: The George Medal was created by King George VI in September 1940 along with the GC to serve as a pair of decorations to recognize civilian and military gallantry in non-combat situations. A number of awards were made to Canadians for their services in defusing bombs placed by the Front de libération du Québec (FLQ). The modern Canadian equivalent is the Star of Courage.
Criteria: Intended as an award primarily for civilians and as an award for the military for "acts of great bravery" for which purely military honours are not normally granted.
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing the Sovereign's effigy and titles, while the reverse displays a representation of St. George slaying the dragon on the coast of England circumscribed by the text THE GEORGE MEDAL. The medal was designed by K.C. Gray. Four obverse types of the George Medal have been awarded to Canadians:
Queen Elizabeth II investing Flight Lieutenant Robert Sabourin with the George Medal, Rideau Hall, 1959.
* • George VI 1st type (1940–49): George VI crowned, circumscribed by the legend GEORGIVS VI D:G:BR:OMN:REX ET INDIAE IMP:.
* • George VI 2nd type (1949–52): George VI crowned, circumscribed by the legend GEORGIVS VI DEI GRA BRITT.OMN REX FID.DEF:+.
* • Elizabeth II 1st type (1952–59): Elizabeth II crowned, circumscribed by the legend ELIZABETH II D:G: BR: OMN: REGINA F:D:.
* • Elizabeth II 2nd type (1959 to present): Elizabeth II crowned, circumscribed by the legend ELIZABETH II DEI GRATIA REGINA F.D..
George Medal George VI 1st type obverse.
Suspender: A footed suspender through which a 15 mm ring passes and the ribbon.
Ribbon: A red ribbon 32 mm in width bearing 5 equally spaced 2 mm wide stripes of garter blue.
Bars: A silver bar 33 mm wide displaying laurel leaves in the centre slides over the ribbon.
Naming: Engraved or impressed capitals around the rim. Various details have been found engraved around the edge; civilian issues are found with both the full given name and surname or initials and surname, while those to members of the armed forces are found with the number, abbreviated rank, initials, surname, and abbreviated service/corps.
Postnominals: GM.
Other: One award of the GM to a Canadian was cancelled at the request of the recipient.
Number of Awards to Canadians: 76(4 1st bars, 1 cancelled award).
* • Second World War: 42 (4 1st bars).
* • Korean War: 1.
* • Postwar: 31 (1 of these awards was cancelled).
The Distinguished Service Medal
Distinguished Service Medal George VI 1st type obverse.
Origins: The Distinguished Service Medal was created by King George V in October 1914 to be an award junior to the Conspicuous Gallantry Medal for members of the naval services other than commissioned officers. A bar to the DSM was created in 1916, and in 1942 the award was opened to those serving in the Merchant Navy. The modern Canadian equivalent is the Medal of Military Valour.
Criteria: Chief petty officers, petty officers, and men of the naval services who distinguish themselves by acts of pre-eminent bravery in action with the enemy.
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing the effigy of the reigning Sovereign circumscribed by his or her titles. Three issues of this medal have been awarded to Canadians:
* • George VI 1st type: Crowned effigy of George VI circumscribed by the legend GEORGIVS VI D: G:BR:OMN:REX ET INDIAE IMP:.
* • George VI 2nd type: Crowned effigy of George VI circumscribed by the legend GEORGIVS VI DEI GRA: BRITT: OMN: REX FID: DEF:+.
* • Elizabeth II: Crowned effigy of Elizabeth II circumscribed by the legend ELIZABETH II D:G:BR:OMN:REGINA F:D:.
Distinguished Service Medal reverse.
The reverse of this medal has remained unchanged since its inception: a Victorian Crown and the text FOR DISTINGUISHED SERVICE on three lines within a laurel wreath.
Suspender: A swivelling claw-footed suspender with a straight bar.
Ribbon: A navy blue ribbon 32 mm wide with a 12 mm stripe of white down the centre divvied in half by a 2 mm wide stripe of navy blue.
Bars: A silver bar 33 mm wide with raised edge displaying laurel leaves in the centre slides on the ribbon.
Naming: Engraved or impressed capitals, number, abbreviated rank, initials, surname, and service.
Postnominals: DSM.
Other: A 2nd type of Queen Elizabeth II obverse also exists, though none were awarded to Canadians. In 1993 the British replaced this decoration with the Distinguished Service Cross.
Number of Awards to Canadians: 116 (2 1st bars).
* • Second World War: 114 (2 1st bars).
* • Korean War: 2.
The Military Medal
Military Medal George V obverse.
Military Medal George V reverse.
Origins: The Military Medal was created in March 1916 to recognize warrant officers, non-commissioned officers, and men serving in the army. Royal Flying Corps personnel and members of the naval services were also eligible for this medal under certain circumstances. The modern Canadian equivalent is the Medal of Military Valour.
Criteria: Acts of bravery in the field.
Insignia: A circular silver medal 36 mm in diameter, there have been six obverse types awarded to Canadians:
* • George V 1st type: George V uncrowned in a field marshal's uniform circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND:IMP:.
* • George V 2nd type: George V crowned in coronation robes circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND:IMP:.
* • George VI 1st type: George VI crowned, circumscribed by the legend GEORGIVS VI D:G: BR: OMN: REX ET INDIAE IMP:.
* • George VI 2nd type: George VI crowned, circumscribed by the legend GEORGIVS VI DEI GRA: BRITT. OMN: REX FID: DEF:+.
* • Elizabeth II: Elizabeth II crowned, circumscribed by the legend ELIZABETH II D:G: BR: OMN: REGINA F:D:.
The reverse displays the Royal Cypher of the reigning Sovereign below which is found the text FOR BRAVERY IN THE FIELD on four lines, the text and cypher within a wreath of laurel leaves. Four reverse type medals were awarded to Canadians:
Military Medal George VI 1st type obverse.
Military Medal George VI 1st type reverse.
* • George V both types: GVR.
* • George VI 1st type: GRI.
* • George VI 2nd type: GVIR.
* • Elizabeth II: EIIR.
Suspender: Swivelling claw-footed suspender with an ornate scroll bar.
Ribbon: A dark blue ribbon 32 mm in width displaying in the centre alternating 3 mm stripes of white (3) and red (2).
Bars: A silver bar with raised edge displaying laurel leaves in the centre slides over the ribbon.
Naming: Engraved or impressed capitals, number, abbreviated rank, initials, surname, and unit.
Postnominals: MM.
Other: In 1993 the British replaced this decoration with the Military Cross.
Number of Awards to Canadians: 13,654 (848 1st bars, 3 2nd bars).
* • First World War: 12,345 (838 1st bars, 37 2nd bars).
* • Second World War: 1,256 (10 1st bars, 1 2nd bar).
* • Korean War: 53.
The Distinguished Flying Medal
Distinguished Flying Medal reverse.
Distinguished Flying Medal George VI 1st type obverse.
Origins: The Distinguished Flying Medal is part of the family of air force decorations created in June 1918 after the establishment of the Royal Air Force.
Criteria: Awarded to non-commissioned officers and men who perform acts of valour, courage, or devotion to duty while flying in active operations against the enemy. The modern Canadian equivalent is the Medal of Military Valour.
Insignia: A silver oval-shaped medal 35 mm wide and 41 mm tall displaying in the centre the Sovereign's effigy circumscribed by his or her titles. The reverse displays a representation of Athena Nike seated on an airplane with the date 1918 in the field and the text FOR COURAGE, the entire allegory encircled within a narrow laurel wreath at the outermost edge of the disc. Two types of this medal were issued to Canadians:
* • George V issue: an uncrowned effigy of George V circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND:IMP:.
* • George VI issue: an uncrowned effigy of George VI circumscribed by the legend GEORGIVS VI D:G: BR: OMN: REX F.D: IND:IMP:.
Suspender: A pair of extended wings joined in the centre by a bomb in the style of the First World War.
Ribbon: A white ribbon 32 mm wide displaying purple stripes on a 45 degree angle, 2 mm apart. Until 1919 the stripes were horizontal.
Bars: A 33 mm wide silver bar with a flying eagle centred; the bar slips over the ribbon.
Postnominals: DFM.
Naming: Impressed around the rim, number, rank, surname, and given initials, or hand-engraved around the rim number, rank, given initials, surname, and RCAF.
Other: One other George V type (crowned), one other George VI type (without IND IMP), and another Elizabeth II type of this decoration also exist; however, none of these were bestowed upon Canadians. In 1993 the British replaced this decoration with the Distinguished Flying Cross.
Number of Awards to Canadians: 516 (all awards for the Second World War).
The Air Force Medal
Air Force Medal George VI 1st type obverse.
Air Force Medal reverse.
Origins: The Air Force Medal was part of the family of air force decorations created in June 1918 after the establishment of the Royal Air Force.
Criteria: Awarded to non-commissioned officers and men who perform acts of valour, courage, or devotion to duty while flying, though not in active operations against the enemy. The modern Canadian equivalent is the Meritorious Service Cross or Medal (military division).
Insignia: A silver oval-shaped medal 35 mm wide and 41 mm tall, displaying in the centre the Sovereign's effigy circumscribed by his or her titles. The reverse displays a representation of Hermes mounted on a hawk in flight bestowing a wreath, along with the date 1918 in the field, the entire allegory encircled within a narrow laurel wreath at the outermost edge of the disc. Two types of this medal were issued to Canadians:
* • George VI issue: uncrowned effigy of George VI circumscribed by the legend GEORGIVS VI D:G: BR: OMN: REX F.D: IND:IMP:.
* • Elizabeth II issue: uncrowned effigy of Elizabeth II circumscribed by the legend ELIZABETH II DEI GRATIA REGINA FID: DEF:.
Suspender: A pair of extended wings joined in the centre by a bomb in the style of the First World War.
Ribbon: A white ribbon 32 mm wide displaying red stripes on a 45 degree angle, 2 mm apart. Until 1919 the stripes were horizontal.
Bars: A 33 mm wide silver bar with a flying eagle centred; the bar slips over the ribbon.
Naming: Impressed around the rim, number, rank, surname, and given initials, or hand-engraved around the rim number, rank, given initials, surname, and RCAF.
Postnominals: AFM.
Other: Two King George V types, one other King George VI type (without IND IMP), and another Elizabeth II type of this decoration also exist; however, none of these were bestowed upon Canadians. In 1993 the British replaced this decoration with the Air Force Cross.
Number of Awards to Canadians: 46.
* • Second World War: 42.
* • Postwar: 2.
* • Korean War: 2.
The Edward Medal in Silver and the Edward Medal
Edward Medal in silver Edward VII obverse.
Origins: The Edward Medal was established by King Edward VII in 1908 to recognize "the many heroic acts performed by Miners and Quarrymen and others who endanger their own lives in saving or endeavouring to save the lives of others from the perils in Mines and Quarries."[5] The medal was unusual in that it came in two classes, 1st and 2nd, and by 1909 the criteria had been expanded to include heroism displayed in the course of industrial employment. In effect, the medal had two divisions — one for mines and one for industry with two classes of recognition for each. By 1917 the designations of 1st and 2nd class were dropped, the 1st class becoming the Edward Medal in Silver and the 2nd class simply as the Edward Medal.
The last award of the Edward Medal in Silver to a Canadian was made in 1916, and it was discontinued in 1917. The Edward Medal was only awarded posthumously from 1949 until it was discontinued in 1971. The modern Canadian equivalent of the Edward Medal in Silver is the Star of Courage, while the Medal of Bravery is the modern equivalent of the Edward Medal.
Criteria: The criteria varied depending on the setting. For Mines and Quarries the medal was bestowed to recognize those saving or endeavouring to save lives in mines and quarries. For Industrial Employment the medal was bestowed to recognize those saving or endeavouring to save lives in an industrial setting. Unlike the Albert Medals, there was no requirement that the probability of death exceed the likelihood of survival.
Edward Medal in silver mines reverse.
Insignia: For the Edward Medal in Silver, two types of obverses were issued to Canadians who earned this medal; both were a circular silver medal 36 mm in diameter:
* • Uncrowned effigy of Edward VII circumscribed by the legend EDWARDVS VII D: G: BRITT: OMN: REX F:D: IND:IMP:.
* • Uncrowned effigy of George V circumscribed by the legend GEORGIVS V D.G.BRITT: OMN: REX F.D. IND:IMP:.
For the Edward Medal, only one type of obverse was issued to Canadians: a circular bronze medal 36 mm in diameter bearing the uncrowned effigy of King George V and circumscribed by the legend GEORGVS V D.G.BRITT: OMN: REX F.D. IND:IMP.
There were three reverse types awarded to Canadians, which was dependent on whether the medal was awarded for Mines and Quarries or for Industrial Employment:
* • Mines and Quarries depicts a miner rescuing a comrade in a collapsed mine shaft with the text FOR COURAGE near the upper left.
* • 1st type Industrial Employment depicts the upper torso of a bare-chested man with his left arm reaching up for help, and in his right arm a limp companion. On the left side an industrial town is depicted, and the upper right carries the text FOR COURAGE.
* • 2nd type Industrial Employment displays a female figure holding a wreath in each of her outstretched hands, with an industrial town in the background and the text FOR COURAGE.
Edward Medal in bronze George V reverse.
Edward Medal in bronze, industry reverse.
Suspender: Claw-footed suspender with an oval ring 9 mm by 14 mm.
Ribbon: A dark blue ribbon 35 mm wide edged on each side by 5 mm of yellow. The same ribbon was used for both honours.
Bars: A bar could be awarded with this medal, though none were awarded to Canadians.
Naming: Hand-engraved capitals, given name, and surname in full.
Postnominals: EM.
Other: In 1971 the living recipients of the Edward Medal were invited to exchange their medals for the George Cross. There were no Canadian exchange awards. Only the versions of those medals earned by Canadians are listed herein.
Number of Awards to Canadians: 5.
* • 1 Edward Medal in Silver (Mines and Quarries).
* • 1 Edward Medal in Silver (Industry Employment).
* • 3 Edward Medals (Industry Employment).
The King's Police Medal
King's Police Medal George V 1st type obverse.
Origins: As the police services throughout the British Empire became increasingly professionalized and important in the maintenance of peace and order, it became necessary to institute an award to recognize the brave acts and/or distinguished services rendered by keepers of the King's peace. While those serving in the various constabularies were eligible for such decorations as the Albert Medal for gallantry, there were no official honours to recognize distinguished service. The Royal Irish Constabulary had established the Constabulary Medal in 1842, but it was limited to service in Ireland. In Canada some members of the Royal Northwest Mounted Police (RNWMP) and the Dominion Police were awarded the Imperial Service Medal, while a very few senior officers were appointed to the Imperial Service Order. However, the breadth of recognition available to the police was very limited.
In 1909 King Edward VII established the King's Police Medal, which could be awarded for gallantry or for distinguished services. The name of the decoration was changed to the King's Police and Fire Services Medal in 1940, though members of the fire services had been eligible for the medal from its inception.
Beginning in 1950, the medal for gallantry was only awarded posthumously, the George Medal being better suited for this sort of recognition. In 1954 this medal was restructured into two separate awards — the Queen's Police Medal and the Queen's Fire Services Medal — none of which were ever awarded on the advice of the Canadian government, though the offer was extended by the British government. The modern Canadian equivalent for KPMs awarded for gallantry is the Medal of Bravery, and for distinguished service the equivalent is an appointment to the Order of Merit of the Police Forces or an awarding of a Meritorious Service Decoration.
Criteria: To recognize those in the police and fire services for heroic acts of courage and instances of conspicuous devotion to duty. Specifically:
King's Police Medal distinguished service reverse.
> a) Conspicuous gallantry in saving life and property, or in preventing crime or arresting criminals; the risks incurred to be estimated with due regard to the obligations and duties of the officer concerned.
>
> b) An especially distinguished record in administrative or detective service.
>
> c) Success in organizing police forces or fire departments, or in maintaining their organ-ization under special difficulties.
>
> d) Special services in dealing with serious or widespread outbreaks of crime or public disorder, or fire.
>
> e) Valuable political and secret services.
>
> f) Special services to Royalty and Heads of State.
>
> g) Prolonged service, but only when distinguished by very exceptional ability and merit.[6]
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing an effigy of the reigning sovereign. There have been three obverse types awarded to Canadians:
* • George V 1st type: An uncrowned George V circumscribed by the legend GEORGIVS V BRITT: OMN: REX ET IND: IMP:.
* • George VI 1st type: A crowned effigy of George VI circumscribed by the legend GEORGIVS VI D: G:BR: OMN: REX ET INDIAE IMP:.
* • George VI 2nd type: A crowned effigy of George VI circumscribed by the legend GEORGIVS VI DEI GRA: BRITT: OMN: REX FID: DEF:+.
There are two distinctive reverses for this medal, one for gallantry the other for distinguished service:
* • When awarded for distinguished service: a helmeted knight standing erect, holding close to his body the sword of justice, his other hand atop a shield bearing the text TO GUARD MY PEOPLE on three lines. In the background of this allegory is a fortified city.
* • When awarded for gallantry, the reverse is identical to that for distinguished service save the addition of the text FOR GALLANTRY at the bottom of the reverse. This reverse was introduced in 1933 and issued in 1934.
Suspender: A claw-footed suspender with an oval ring 9 mm by 14 mm.
Ribbon: Three ribbon types were used with this medal:
* • A blue ribbon 35 mm wide edged on each side with 4 mm of white (1912–16).
* • A blue ribbon 35 mm wide with a central 4 mm wide stripe of white and edged on each side by 4 mm of white (1916–33, for both distinguished service and gallantry; from 1933 onward only for distinguished service).
* • A blue ribbon 35 mm wide, with a central 4 mm stripe of white and edged on each side by 4 mm of white; in the centre of each white stripe is 1 mm of red (to denote the medal was awarded for gallantry, used from 1933 onward).
Bars: A silver laurelled bar, though none awarded to Canadians.
Naming: Engraved capitals, most often found with the abbreviated rank, full given and surname of the recipient. The name of the police service is also occasionally included.
Postnominals: KPM or KPFSM (not introduced until 1969).
Other: A King Edward VII, a second King George V type, and two types of Queen Elizabeth II issues of the medal also exist, though none were awarded to Canadians.
Number of Awards to Canadians: 50 (8 George V 1st type, 37 George VI 1st type, 5 George VI 2nd type).
Mention in Dispatches
Mention in Dispatches oak leaf First World War issue.
Mention in Dispatches oak leaf, Second World War and Korea War issue.
Origins: Prior to the creation of official gallantry decorations, it was customary to mention individuals who had distinguished themselves in the field or at sea in dispatches. These dispatches or reports were sent by commanders to the Sovereign and to Parliament to inform them of progress in a particular campaign. To be mentioned in and of itself was an honour, though it was not originally accompanied by an insignia or certificate. The tradition originated in the Royal Navy as a mechanism for allowing flag officers to learn of the service and actions of naval officers serving at sea. Exemplary action would be noted and placed in an officer's record and was, in the long term, often crucial in securing promotions. As the British Empire expanded, it was no longer just the navy that saw action far away from the British Isles but also the army. Thus the practice spread to the land forces, though it would not be until 1844 that it was extended to non-commissioned officers.[7]
Mentions in Dispatches (MIDs) became widely used in the South African War (1899–1902), and a handful of Canadians were awarded them. Prior to the establishment of the broad range of British gallantry awards, there was no real intermediate award for bravery in the field that did not warrant either the Distinguished Conduct Medal, Distinguished Service Order, or Victoria Cross, and the MID filled this gap.
During the First World War, Canadians continued to receive MIDs. Because the British honours system quickly expanded in this period to include a variety of intermediate bravery awards, MIDs were not greatly used by the Canadian Expeditionary Force. Also because of this expansion, the MID lost some of its cachet as an award of much significance: soldiers would much rather receive a medal than a simple mention in the London Gazette. Indeed, more Canadians received the Military Medal than were awarded an MID.
Military personnel in an MID had their names listed in the London Gazette and received certificates from Britain's secretary of state for war, along with small oak leaf branches for wear on their Victory Medals (created in 1920). This practice continued during the Second World War and again during the Korean War, after which the practice went into abeyance after the establishment of the Canadian honours system.
First World War Mention in Dispatches certificate.
Criteria: For valiant conduct, devotion to duty, or other distinguished service in warlike situations in an active theatre of operations.
Insignia: First introduced in 1920, with permission to wear insignia backdated to those recognized beginning in 1914. There have been two types of full-sized MID insignia:
* • 1st type (1914–20), a multi-lobed oak leaf, bronze in colour, 36 mm in length.
* • 2nd type (1920 to present), a single oak leaf, bronze in colour, 20 mm in length.
Wear: The insignia is worn on the designated ribbon (Allied Victory Medal, 1939–45 War Medal, Queen's Korean War Medal). If there is no designated ribbon, the insignia is worn at the end of the undress ribbons sewn to the uniform.
Other: Beginning in 1919, recipients were presented with a certificate bearing the Royal Arms of the United Kingdom at the top along with their rank, name, service, and serial number (where applicable).
Second World War Mention in Dispatches certificate.
Number Awarded: 5,467 (First World War), 9,666 (Second World War), 260 (Korean War).
King's/Queen's Commendation for Brave Conduct
King's/Queen's Commendation for Brave Conduct badge.
Origins: Created primarily to serve as a civilian equivalent to a Mention in Dispatches, the first awards of this commendation were made in December 1939. During the First World War, various civilians were commended officially by the British government. However, there was no consistency to the award, nor was there an insignia of any type.
Criteria: Awarded for brave conduct worthy of recognition but not meeting the requirement for bestowal of a gallantry decoration such as the George Medal.
Insignia: For those serving in a uniformed service, a Mention in Dispatches oak leaf was awarded. Civilians received an oval-shaped, gold-coloured badge 22 mm wide and 40 mm tall (initially made of plastic!) displaying two laurel braches terminating with the Royal Crown at the top, in the centre a broad sword and a plaque bearing the text FOR BRAVE CONDUCT. After the Second World War, this badge was replaced with a silver spray of laurel leaves worn on the ribbon of the Defence Medal or on the lapel in the absence of an associated medal.
Postnominals: None.
Number of Awards to Canadians: 28 King's and 96 Queen's Commendations.
King's/Queen's Commendation for Brave Conduct ribbon insignia.
King's/Queen's Commendation for Valuable Service in the Air
King's/Queen's Commendation for Valuable Service in the Air.
Origins: The King's/Queen's Commendation for Valuable Service in the Air was created in 1943.
Criteria: Awarded for brave conduct worthy of recognition but not meeting the requirement for bestowal of a gallantry decoration such as the Air Force Cross or Air Force Medal.
Insignia: A bronze Mention in Dispatches oak leaf was worn on the relevant war service medal ribbon or on the lapel in the absence of an associated medal. Civilians awarded this commendation received a silver-coloured badge composed of a pair of stretched wings with the text FOR VALUABLE SERVICE atop an Imperial Crown.
Postnominals: None.
Number of Awards to Canadians: 319.
# 6
IMPERIAL CAMPAIGN MEDALS
Although medals commemorating service in battle are known to date back to the time of Elizabeth I and the defeat of the Spanish Armada, tthese awards were not intended for wear. The first broadly distributed war medal intended for wear by both officers and men was the Waterloo Medal issued in 1815, which set the pattern for the various service medals awarded throughout the Commonwealth to this day. The British Naval General Service Medal and Military General Service Medal, both established in 1847, served as models for the multi-engagement war and operational service medals that came after. A few earlier issues, such as the Sutlej Medal, are of a similar design, but the former two medals were the most widely awarded.
The Naval and Military General Service Medals, though not awarded solely for service in North America, were the first modern-style medals awarded to Canadians or to others for service in what would become Canada. The first British medal awarded for service solely in Canada was the North West Canada Medal, instituted in 1885 for those who had served in quelling the North-West Rebellion. The obverse of the medal displayed a youthful Queen Victoria, while the reverse featured a wreath of maple leaves and the inscription NORTH WEST CANADA 1885. It was an austere design, but it set a precedent in Canada as the first true Canadian military service medal. It was largely a Canadian expedition, one in which few British officials were involved.[1]
The striking of the medal was initiated by Lord Lansdowne, Canada's governor general. On May 19, 1885, Lansdowne suggested to the British government that a medal recognizing service in the rebellion should be struck, since it "would have an excellent effect upon the spirit of the Canadian forces."
Just over a month later, on June 22, 1885, the British government agreed that such a medal could be struck on the condition that the cost was borne by the Canadian government.[2] Lansdowne was elated with this news, though much less satisfied that the Canadian government was to be saddled with the bill. He viewed the cost as incidental, given the "results which would follow from the free gift of the medal." It was ultimately agreed that the imperial government would pay for the honour.
The governor general even took an interest in the design: "The obverse will, I assume, represent the Queen's head, and the reverse might be engraved with a design distinctive of the Dominion."[3] Lansdowne later suggested that the reverse of the medal "should be ornamented with a wreath of Maple leaves, in the centre of which the words 'The North West 1885' might be impressed."[4] While the governor general developed the concept, it was Allan Wyon of the Royal Mint who completed the formal design.
The Canadian government, never having been involved with creating a military medal, briefly considered having the medal struck in bronze, though Lansdowne insisted that silver be used. They were unaware that bronze medals were usually reserved for non-European troops serving in India.
An order-in-council was issued on February 16, 1887, after Queen Victoria authorized the design and striking of the medal. With this Canada's first military medal was born. The design and purpose of such service medals has remained largely the same since the striking of the North West Canada Medal with minor variations. Even the eighteen-month period that it took to get the medal approved is comparable to the time it takes to have a modern Canadian award developed from a rough concept to a completed design.
The next medal with a connection to Canada was the aptly named Canada General Service Medal, sanctioned in 1899 by Queen Victoria for those who had served in the 1866 and 1870 Fenian Raids as well as in the 1870 Red River Rebellion, Riel's first uprising. The obverse depicted Queen Victoria, while the reverse contained a spreading wreath of maple leaves and the Canadian Red Ensign. Most significantly, the ribbon comprised three equal stripes: red, white, and red, possibly taken from the flag used by the Royal Military College of Canada, founded in 1876.
The Canada General Service medal inaugurated another tradition, the creation of medals many decades after the actual operation or event, there having been a thirty-three year gap between the first Fenian Raid and the issue of the medal.
Chapter 23 provides a detailed account of Canada's post-1967 war and service medals, while this chapter focuses on those awarded prior to the creation of the modern Canadian honours system. Each of the following entries includes a short description of the conflict in question: these are by no means complete accounts of each war, battle, or engagement; however, they are intended to provide some background to the events recognized with each medal and or clasp.
The Military General Service Medal, 1793–1814
Military General Service Medal obverse.
Origins: The Military General Service Medal was instituted on June 1, 1847, to recognize officers and men who served in a diverse array of wars and engagements that took place between 1793 and 1850. Of the twenty-nine clasps awarded with this medal, twenty-one were awarded for service in the Peninsular War (1808–14) and three were awarded for service in the War of 1812, specifically for engagements at Fort Detroit, Chateauguay, and Crysler's Farm. It is the three War of 1812 clasps and battles that make this medal of great interest to Canadians. An attempt was made to have clasps created for other key battles, namely, Queenston Heights, Lundy's Lane, Ogdensburg, Lacolle, and Plattsburg. However, the War Office refused.
Hostilities began on June 18, 1812, when the United States declared war on Britain — and thus Britain's North American colonies were drawn into the fray. Tensions between Britain and the United States had grown as a result of events surrounding the Napoleonic Wars, one of the factors being Napoleon's closing of all European ports to British trade. In response to this action, the British began requiring that all neutral ships sailing to Europe have a licence, which had a profound effect on American trade interests, since the United States had maintained neutrality during the conflict.
The British routinely stopped American vessels on the high seas and searched for contraband and deserters from the Royal Navy who had joined U.S. ships. While many of the deserters had since become American citizens, this meant little to the British, who would arrest them and re-impress them into the Royal Navy. While the British government rescinded the orders-in-council permitting this sort of action a few days before the American declaration of war, the slowness of communication meant the news did not reach Washington until after the declaration had been made. The war was fought in five theatres: Atlantic, Western, Niagara, St. Lawrence, and Southern.
The Atlantic operations transpired throughout the entire war period. The Royal Navy, based out of Halifax and the West Indies, maintained a blockade of American ports in an effort to restrict the flow of trade and goods. The most significant naval engagement of the war took place in September 1813 between the HMS Shannon and the USS Chesapeake, which resulted in a humiliating defeat for the Americans outside Boston's harbour. In addition to this, ships primarily from Nova Scotia were provided with letters of marque that allowed them to legally capture and requisition enemy ships and goods. The Royal Navy also launched attacks on American towns along the Eastern Seaboard.
Military General Service Medal with multiple bars obverse.
Western operations took place in the Great Lakes region, largely over all-important transportation routes. Notable among these operations was the capture of Fort Detroit by British and Canadian troops aided by First Nations warriors. Other incursions included attacks on Fort Dearborn (Chicago), the capture of the post at Michilimackinac, and the final battle at Moraviantown in October 1813.
The Niagara operation involved the many forts and strongholds on the Canadian and American sides of the Niagara River. Almost a dozen battles were fought from Niagara to what is Toronto today. The burning of York (Toronto), the Battle of Queenstown Heights, and the Battle of Beaver Dams were key engagements. The St. Lawrence operation included an attempt by American troops to capture Montreal in an effort to prevent British reinforcements from reaching the inter-ior of Upper Canada and Niagara. The Great Lakes and Lake Champlain witnessed fierce fighting on the water between British/Canadian and American forces. The conflict also included Southern operations such as the Creek War and the Battle of New Orleans.
The Treaty of Ghent ending the war was signed by British and American representatives on December 24, 1814, though it would be several months until the hostilities came to an end on February 17, 1815.
Criteria: Participation in any number of battles or campaigns. This medal could be awarded to a diverse array of combatants, not only those serving in the British Army, Canadian Militia, or Royal Newfoundland Fencibles, but also First Nations warriors. Applications for this medal could only be made by survivors.
Military General Service Medal reverse.
Insignia: A circular silver medal 36 mm in diameter, depicting on the obverse an effigy of Queen Victoria circumscribed by the legend VICTORIA REGINA with the date 1848 at the base. The reverse depicts Queen Victoria standing atop a dais and presenting a kneeling Duke of Wellington with a laurel wreath. Beside the dais is an allegorical British lion. The entire scene is circumscribed by the text TO THE BRITISH ARMY with the dates 1793–1814 in the exergue. The medal was always awarded with at least one bar and was designed by William Wyon.
Suspender: A swivelling claw suspender with a straight bar to which clasps attach.
Ribbon: Crimson in colour, 31 mm wide, edged with 3 mm of dark blue on each side.
Clasps: Twenty-nine battle/campaign clasps were issued, and it was possible for a soldier or officer to receive multiple clasps, the maximum awarded being fifteen. The Canadian-related clasps awarded for the War of 1812 were:
* • FORT DETROIT: Sir Isaac Brock, the Canadian Militia, and a large group of First Nations warriors captured the fort from a vastly superior American force on August 16, 1813.
* • CHATEAUGUAY: Lieutenant-Colonel Charles de Salaberry and those under his command who defended Montreal from the advancing U.S. force of Major-General Wade Hampton. Aside from a few members of the Royal Artillery, this battle was fought almost entirely by members of the Canadian Militia and First Nations warriors on October 26, 1813.
* • CRYSLER'S FARM: The U.S. Army under General James Wilkinson planned another attack on Montreal that was to involve Major-General Hampton and his forces. However, Hampton failed to join the attack. A force led by Lieutenant-Colonel Joseph W. Morrison was ordered to attack the rear of the American force. However, Wilkinson learned of this and ordered an attack upon Morrison's force, which had taken up a position at Crysler's Farm where the battle took place on November 11, 1813.
Naming: Impressed Roman capitals with given name and surname for men, while officers' medals included their abbreviated rank.
Other: A number of unofficial clasps for Stoney Creek, Fort George, and Queenstown Heights are known to exist.
Number: 908.
* • 315 Fort Detroit.
* • 339 Chateauguay.
* • 211 Crysler's Farm.
* • 3 Fort Detroit and Chateauguay.
* • 8 Fort Detroit and Crysler's Farm.
* • 3 with all 3 clasps.
* • 7 Fort Detroit with other non-Canada–related clasps.
* • 1 with Chateauguay and other non-Canada–related clasp bars.
* • 21 with other non-Canada-related clasps.
The Naval General Service Medal, 1793–1840
Naval General Service Medal obverse.
Origins: The Naval General Service Medal was instituted in 1847 to recognize officers and men who had served in the Royal Navy and Royal Marines who had been in a diverse array of wars and engagements that took place between 1793 and 1840, notably during the French Revolutionary War and Napoleonic Wars. A total of 231 different clasps were issued with this medal, and the maximum awarded with a single medal was seven. Eleven clasps were issued in relation to War of 1812 actions.
Criteria: Participation in any number of actions. Applications for this medal could only be made by survivors. The medal was always issued with at least one clasp. Those most relevant to Canada were awarded to those who served in boat service engagements that took place on the Great Lakes during the War of 1812 on September 3 and 6 and May 6, 1814.
Insignia: A circular silver medal 36 mm in diameter, depicting on the obverse an effigy of Queen Victoria circumscribed by the legend VICTORIA REGINA with the date 1848 at the base. The reverse depicts Britannia triumphally holding a trident while astride a seahorse. The medal was designed by William Wyon.
Suspender: A swivelling claw suspender with a straight bar to which clasps attach.
Ribbon: White in colour, 31 mm wide, edged with 3 mm of dark blue on each side.
Clasps: There are 231 different clasps for everything from major battles, minor engagements, and boat service actions. Those related to the War of 1812 are:
* • 28 APRIL BOAT SERVICE 1813: HMS Dolphin, HMS Dragon, HMS Fantome, HMS Highflyer, HMS Maidstone, HMS Marlborough, HMS Mohawk, HMS Racer, and HMS Statira sailed up the Elk River from Chesapeake Bay with the objective of destroying American ships and the cannon foundry at Frenchtown. After destroying five American ships and Frenchtown on April 29, a landing party proceeded to attack the town of Havre de Grace on May 3. Clasps were given for two separate actions. Fifty-five clasps were issued.
Naval General Service Medal reverse.
Boat Service Bar for the Naval General Service Medal.
* • 29 APRIL BOAT SERVICE 1813: HMS Orpheus burned the Whampoa, an American merchant vessel, in Narragansett Bay off Rhode Island. Two clasps were issued.
* • PELICAN: HMS Pelican captured the USS Argus off the coast of Ireland on August 14, 1813. The Argus had been raiding British shipping in the Bristol Channel area. One hundred and eleven clasps were issued.
* • SHANNON WH. CHESAPEAKE: Capture of the USS Chesapeake by the HMS Shannon off the coast of New England on June 1, 1813. Forty-eight clasps were issued.
* • PHOEBE: HMS Phoebe engaged USS Essex and captured it along with the sloop Essex Junior off the coast of Chile on March 28, 1814. Thirty clasps were issued.
* • CHERUB: HMS Cherub engaged USS Essex and captured it along with the sloop Essex Junior off the coast of Chile on March 28, 1814. Nine clasps were issued.
* • 8 APRIL BOAT SERVICE 1814: HMS Hogue, HMS Edymion, HMS Maidstone, and HMS Boxer destroyed American vessels in the Connecticut River. Twenty-four clasps were issued.
* • THE POTOMAC 17 AUGUST 1814: HMS Aetna, HMS Devastation, HMS Euryalus, HMS Leviathan, HMS Meteor, HMS Sea Horse, HMS Erebus, HMS Fairy, HMS Anna Maria, HMS Alceste, and HMS Regulus for the destruction of American vessels 140 kilometres up the Potomac River and the bombardment and capture of Fort Washington and Alexandria, Virginia, from August 17 to September 6, 1814. One hundred and eleven clasps were issued.
* • 3 & 6 SEPT. BOAT SERVICE 1814: HMS Nancy aided by the Royal Newfoundland Regiment destroyed the American schooners USS Tigress and USS Scorpion on Lake Huron. One bar was issued to Lieutenant Andrew Bulger, who also earned the Military General Service Medal with clasps for Fort Detroit and Crysler's Farm.
* • 6TH MAY BOAT SERVICE 1814: Aided by the army, a contingent of two hundred men landed via boats and attacked Fort Orivego on the shore of Lake Ontario. One clasp was issued to James Wills.
* • 14TH DEC. BOAT SERVICE 1814: Awarded for the Battle of Lake Borgne near New Orleans. This action resulted in the capture of five American gunboats and a sloop. Two hundred and fourteen clasps were issued.
* • ENDYMION WH. PRESIDENT: Toward the end of the War of 1812 the USS President had been blockaded in New York but escaped to be chased by the HMS Endymion, and the two ships engaged off Sandy Hook. However, the President managed to escape only to encounter the HMS Pomone and HMS Tenedos, to which the President surrendered after a short engagement on January 15, 1815. Sixty-three clasps were issued.
Naming: Impressed Roman capitals with given name and surname for men, while officers' medals included their abbreviated ranks.
Other: In total 24,000 medals were issued, 20,933 with a single clasp.
Number: Unknown for those related to Canada.
The Canada General Service Medal, 1866–70
Canada General Service Medal obverse.
Canada General Service Medal reverse.
Origins: This medal was created to recognize members of the Canadian Militia and British forces who saw service in Canada during the Fenian Raids of 1866 and 1870 as well as those who participated in the suppression of the Red River Rebellion. The Fenian Raids were precipitated by a group of Irish-American nationalists who sought to invade Canada and hold it for ransom, hoping that would force the British government to grant Ireland independence. Canadian authorities were aware of the Fenians' intentions and mobilized the militia on March 7, 1866, with nearly ten thousand men called out. The militia was placed at public buildings, bridges, and border crossings in preparation for the attack. However, the extensive nature of the border made it impractical to defend it entirely.
The first incursion came at Waterloo, Quebec, on the evening of March 17, followed by an attack on April 14 on the Campobello and Indian Islands near Maine. These engagements were small in nature, with the largest attack coming on May 31 when John O'Neill and a force of 850 Fenians — in part demobilized soldiers who had fought in the U.S. Civil War — crossed the Niagara Frontier to invade Canada. On June 2 the Fenians were met by a force of 840 Canadian Militia troops, and the Battle of Ridgeway ensued for two hours, though the Canadians retired when it was falsely reported that the Fenians were going to attack with cavalry. The Fenians, learning that a larger force was en route, returned to the security of the U.S. side of the border.
A final engagement took place on June 22 when a small party of Fenians crossed the Vermont border into Quebec and fired at a group of seventy-five soldiers present at Pigeon Hill. The Fenians again fled across the border. Matters subsided and the U.S. government condemned the Fenian incursions. O'Neill and the Fenians returned four years later on May 25, 1870, with two raids taking place, the first at Eccles Hill and a second on the Trout River on May 27.
The Red River Expedition took place between May 14 and August 24, 1870, when Colonel Garnet Wolseley's force arrived at Fort Garry (Red River). The origins of the conflict can be found in the expansion of Canadian authority over western Canada and the maltreatment of the Métis and First Nations populations. The government had obtained sovereignty over Rupert's Land from the Hudson's Bay Company in 1869 and embarked on an ambitious plan to survey the mass of territory and to assert political authority, something that became a paramount point of contention.
The Métis were unhappy with plans to do land surveys, particularly since they did not possess clear title to the land and had settled it according to the seigneurial system rather than the township system. As a result, fears grew that their property would be expropriated. The Métis disrupted the surveying work and eventually prevented the new lieutenant governor from entering the territory. Subsequently, they captured Fort Garry on November 2, 1869.
Louis Riel, with the support of many Métis, attempted to negotiate terms with Canada and demanded fourteen specific rights be granted. Following unrest among some of the anglophone population, who were by this point imprisoned for their resistance, the Métis National Committee proclaimed a provisional government. The government sent Donald Smith (later to become Lord Strathcona) to negotiate an agreement, dealing with matters such as bilingualism and representation in the legislative body to be.
Around the same time a number of the anglophones who had been imprisoned had escaped, while others were paroled. Some of this group began agitating to overthrow the provisional government and gathered recruits. A party of nearly fifty was captured by Riel, and a trial was held in which one of the leaders of the anglophone group, Thomas Scott, was convicted of defying the authority of the provisional government and sentenced to death. A botched execution took place on March 4, and even after being shot in the face, Scott managed to survive for an additional ten hours.
The execution of Scott, an avowed Orangeman, galvanized opinion in English and Protestant Canada and played a role in the government's decision to send the military to Red River. The arrival of Wolseley signalled the end of the rebellion. Riel fled and remained in exile until 1884. The rebellion played a central part in the creation of the Province of Manitoba, which entered Confederation in 1870.
Criteria: Service on active duty in the regular and colonial forces during the Fenian Raid in 1866, the Fenian Raid in 1870, or participation in the Red River Expedition in 1870. The service required as set by AO 32/January 1889 specified that the medal would only be awarded to survivors who "during the operations specified... performed under competent authority 1) active service in the field, or 2) served as guards at any point where an attack from the enemy was expected, or 3) who were detailed for some specific service or duty."
Insignia: A silver medal 36 mm in diameter bearing on the obverse a veiled effigy of Queen Victoria circumscribed by the text VICTORIA REGINA ET IMPERATRIX, the reverse displaying a wreath of natural maple leaves around the Canadian Red Ensign with the text CANADA at the top. Obverse design by T. Brock; reverse by George William de Saulles.
Suspender: A swivelling claw suspender with a straight bar.
Ribbon: A 32 mm wide ribbon divided into equal portions of red, white, and red.
Clasps: A clasp measuring 8 mm by 35 mm with a raised edge. Thee clasps were issued with this medal, individually and in various combinations:
* • FENIAN RAIDS 1866.
* • FENIAN RAIDS 1870.
* • RED RIVER 1870.
Naming: Impressed or engraved upper and lower case letters, rank, initials, surname, and unit.
Other: Specimens engraved with CWM SPECIMEN also exist.
Number: 17,644.
* • Fenian Raid 1866: 11,221.
* • Fenian Raid 1870: 4,510.
* • Red River: 355.
* • Fenian Raid 1866 and Fenian Raid 1870: 1,411.
* • Fenian Raid 1866 and Red River 1870: 120.
* • Fenian Raid 1870 and Red River 1870: 15.
* • All Three Clasps: 12.
The North West Canada Medal (1885)
Origins: After the failure of the Red River Rebellion of 1870, the Métis leader Louis Riel went into exile, only to return with much encouragement from the Métis and other settlers in July 1884. The expansion of the Canadian Pacific Railway into the Canadian West and dwindling bison population had the effect of further marginalizing the Métis and First Nations peoples. Despite the repeated entreaties of the Métis and First Nations to secure certain rights and land, the dominion government ignored the situation. In an effort to force the government to negotiate, Riel formed the Provisional Government of Saskatchewan at Batoche.
North West Canada Medal obverse.
The situation escalated rapidly, with the Métis attacking a detachment of the North-West Mounted Police (NWMP) and Prince Albert Volunteers at Duck Lake. Ottawa realized that a full-scale uprising was in the offing. As a result, the militia was mobilized and a force was dispatched to suppress the rebellion. After the Battle of Duck Lake, Battleford was looted by a Cree raiding party, which was followed by the Frog Lake Massacre, where nine locals were killed, including a government official. At Fort Pitt, Cree intercepted a NWMP scouting party, and the town garrison surrendered. The Battle of Fish Creek halted General Middleton's advance on Batoche, the Métis having defeated a much larger force of government troops. Outside Battleford, the Battle of Cut Knife Hill took place, and again government forces were defeated, this time by the Cree.
The key battle of the rebellion took place on May 9, 1885, with the Battle of Batoche where the Métis, greatly outnumbered, exhausted their supply of ammunition and Riel surrendered on May 15. Several smaller engagements followed, namely, the Battle of Frenchman's Butte and the Battle of Loon Lake. The conflict concluded on July 2, 1885, with the surrender of Chief Big Bear.
Criteria: Awarded to officers and soldiers who participated in quelling the North-West Rebellion. Initially, only those members of the North-West Mounted Police (NWMP) who had served under fire were eligible for the medal. The criteria for this medal were gradually broadened with the passage of time and lobbying efforts on the part of various veterans. In 1900 an order-in-council was approved allowing for NWMP members not under fire to receive the medal. The last group to be awarded the medal was civilian members of the Transport Service, which occurred in April 1945. Recipients of the medal were also granted 320 acres of land and script worth $80. The clasp SASKATCHEWAN was awarded to those who participated in any of the three main engagements that took place during the rebellion, notably those at the Saskatchewan River (Cut Knife Hill), Fish River, and Batoche.
North West Canada Medal reverse.
Insignia: A silver medal 36 mm in diameter, the obverse bearing an effigy of Queen Victoria wearing a diadem and veil, circumscribed by the text VICTORIA REGINA ET IMPERATRIX. The reverse displays a flourish of natural maple leaves in a wreath with the text NORTH WEST 1885 CANADA in the centre on three lines. The design concept was devised by Lord Lansdowne, the formal design of the obverse by L.C. Wyon, and of the reverse by Thomas Brock.
Saskatchewan Bar to the North West Canada Medal.
Suspender: A swivelling claw suspender with a straight bar.
Ribbon: A pearl grey ribbon 32 mm wide edged on each side by stripes of crimson inset 3 mm from the outer edge.
Clasps: A silver clasp 35 mm wide and 7 mm high with a double raised edge and the text SASKATCHEWAN. The medal was not always issued with a clasp.
Naming: This medal is found both unnamed and in a variety of naming styles.
Other: An unofficial clasp, BATOCHE, can also be found occasionally. Copies of both clasps are known to have been manufactured.
Number: 5,650 (1,753 with the clasp SASKATCHEWAN).
The Egyptian Medal (1882–89)
Egyptian Medal obverse.
Egyptian Medal reverse.
Origins: With the opening of the Suez Canal in 1869, the strategic and commercial importance of Egypt became greatly enhanced. Egypt was ruled by the Ottoman Empire khedive, Ismail Pasha, who through financial mismanagement and corruption precipitated great political instability in the country that resulted in his removal by the Turks in 1879. That same year, Colonel Ahmad Arabi led a revolt against Tewfik Pasha, Ismail Pasha's son and successor, and for three years the country was embroiled in further turmoil. After a squadron of British and French ships arrived in Alexandria's harbour in 1882 demanding that Tewfik Pasha's authority be restored, anti-European sentiment became intense, riots broke out in the city, and the European population was attacked.
By early June, rebellion broke out and the Egyptian authorities were unable to quell the disturbances or prevent a massacre. With this the British fleet bombarded Alexandria, the French having withdrawn. British forces subsequently landed and sought to restore Tewfik Pasha's power and drive Arabi out of the country, which was achieved after the capture of Cairo in September 1882.
With the conquest of Egypt, Britain inherited Sudan, which was administered by Egypt and had long been a haven for the slave trade. While Britain and Egypt were at war, a religious fanatic known as the Mahdi rose to power in Sudan. By 1884 the British, along with their new Egyptian allies, marched through Sudan to rid the country of the Mahdi. The initial force was massacred with a loss of ten thousand men. Major-General Charles G. Gordon, a British army officer who had formerly served as governor general of Sudan, was placed in command of a force largely composed of the Egyptian Army. By March 1884, Gordon was besieged in Khartoum with no means of escape, and debate raged in Britain about what should be done. After nearly six months of discussion, General Wolseley was ordered to rescue Gordon.
The British government requested that Canada send troops to assist in the Egyptian campaign, an appeal already made to the colonial government in New South Wales (Australia). Prime Minister Sir John A. Macdonald, however, declined to have the Canadian militia involved but did support Wolseley's request for Canadian boatmen to navigate the rapids of the Nile River on specially constructed boats.
Wolseley's experience during the Red River Expedition greatly influenced his confidence in this mode of transport. The Canadians were to pilot the boats 960 kilometres from Aswan to Khartoum. The first flotilla of thirty-two boats departed from Aswan on October 7. A small force lead by Sir Charles Wilson succeeded in landing 240 Egyptian and Sudanese soldiers and twenty British soldiers at Khartoum on January 28, 1885, well ahead of Wolseley's scheduled arrival in mid-March. While Gordon held out for 317 days, he had been killed forty-eight hours before the arrival of the relief force on January 24, 1885. Despite this, Khartoum was taken and the Mahdi fled and died six months later.
Criteria: Awarded to those who participated in the Egyptian campaigns that took place between 1882 and 1889. Medals to Canadians were awarded with the clasp THE NILE 1884–85. Those Canadian boatmen who made it to Khartoum as part of the force sent to relieve General Gordon also received the clasp KIRBEKAN. Two members of the Canadian Artillery were attached to the Royal Artillery, and medals were issued to them with their Canadian unit listed.
Insignia: A silver medal 36 mm in diameter, the obverse bearing an effigy of Queen Victoria wearing a diadem and veil, circumscribed by the text VICTORIA REGINA ET IMPERATRIX. The reverse displays a sphinx with the legend EGYPT above. The obverse was designed by L.C. Wyon, the reverse by John Pinches.
Suspender: A swivelling claw suspender with a straight bar.
Ribbon: A 32 mm wide ribbon made up of alternating equal stripes of blue, white, blue white, and blue.
Clasps: Thirteen clasps were awarded with this medal, though Canadian recipients were only eligible for two:
* • THE NILE 1884–85.
* • KIRBEKAN — always awarded in combination with THE NILE 1884–85.
Naming: All medals to Canadians are named in sloping impressed capitals.
Other: Medals are found for the Caughnawana, Manitoba, Three Rivers, Ottawa, Peterborough, Sherbrooke, and Sydney Contingents.
Number: 377.
The Khedive's Star
Khedive Star obverse.
Origins: Created by Tewfik Pasha, the khedive of Egypt, to recognize those members of the British and Egyptian forces who served in the Egyptian and Sudanese campaigns of 1882–89.
Criteria: Service in the Egyptian and Sudanese campaigns. The medal was awarded on its own to members of the Egyptian naval and military forces, and it was additionally awarded to British doctors, nurses, and press correspondents by the khedive for service in Egypt.
Insignia: A bronze five-pointed star 48 mm wide, the obverse depicting in the centre the Great Sphinx with the three pyramids at Ghiza in the background, the effigy surrounded by a circle bearing the text EGYPT 1884–6, with the same repeated in Arabic at the base. The reverse displays the khedive's cypher and a crown.
Suspender: An ornamental bronze bar bearing a crescent and star in the centre is attached to the top of the main insignia via a small loop and ring.
Ribbon: A dark blue ribbon 38 mm wide.
Clasp: TOKAR, though none to Canadians.
Naming: None.
Other: These medals were made by Henry Jenkins & Sons of Birmingham. Not all of those entitled to this medal received it.
Number: 312.
The South Africa Medal, 1899–1901 (Queen's South Africa Medal)
Queen's South Africa Medal obverse.
Origins: The South African Medal was instituted in 1900 to recognize naval and military service in the South African War, which had commenced in 1899 between the British Empire and the Afrikaner republics in South Africa — the Transvaal (also known as the South African Republic) and the Orange Free State. The Afrikaners, also known as Boers, had settled in southern Africa in the seventeenth and eighteenth centuries after emigrating from the Netherlands. Throughout the late nineteenth century the European powers, notably Britain, France, Portugal, Spain, Italy, and Germany, raced to acquire African colonies, all this despite the fact the continent was already populated. The discovery of gold and diamonds also helped fuel interest in southern Africa.
On October 9, 1899, President Kruger of the Transvaal issued an ultimatum to the British demanding that they withdraw their troops from his republic's border. The British refused to comply, and the Transvaal and the Orange Free State declared war on Britain on October 11 and invaded Cape Colony and Natal. From there the conflict escalated.
Queen's South Africa Medal reverse with raised dates.
The first phase of the war was marked by a series of defeats for the British, culminating in the "Black Week," which came in the middle of December 1899. The second phase, which can be well traced through the clasps awarded with this medal, lasted from February to August 1900 and resulted in a steady progress by British forces to Pretoria and Bloemfontein, capitals of the two South African republics. Even after the fall of Pretoria, the war continued for an additional two years, characterized by guerrilla warfare and raids. Hostilities concluded on May 31, 1902, with the Treaty of Vereeninging. Eight years later to the day, the Union of South Africa was established out of the previously warring republics and colonies.
Canada became involved in the war after a request from the British government for dominion and colonial troops. Initially, a small force of one thousand was dispatched. However, British losses and fervent pro-British feelings in English Canada forced Prime Minister Sir Wilfrid Laurier to increase the number, which by war's end totalled over six thousand. By comparison, the smaller Australian colonies sent fifteen thousand. Eventually, soldiers from every corner of the empire joined the fight: Australia, New Zealand, Ceylon, India, Rhodesia, Canada, and of course Britain's South African colonies.
Criteria: Service in the South African War between October 11, 1899, and May 31, 1902.
Insignia: A silver medal 36 mm in diameter, the obverse depicting a crowned effigy of Queen Victoria circumscribed by the legend VICTORIA REGINA ET IMPERATRIX. The reverse displays Britannia holding the Royal Union Flag in her left hand with a laurel wreath of victory in her right. In the background can be seen troops marching to the coast and two battleships. A small number are found with the raised dates 1899–1900 on the reverse. There were two strikings of this medal. In the first the wreath of Britannia points to the R in Africa, while in the second it points to the F. The medal was designed by George William de Saulles.
Queen's South Africa Medal reverse without raised dates.
Suspender: A swivelling claw-footed suspender with a straight bar.
Ribbon: The ribbon is 32 mm in width bearing a central stripe of orange bordered on each side by stripes of dark blue and red, each being 5 mm in width.
Clasps: Twenty-six clasps were authorized, though only ten of these are found on medals to Canadians. Each clasp is 3.5 mm by 35 mm and has a raised edge. The maximum number of bars on a single medal was nine.
* • CAPE COLONY: Service in Cape Colony between October 11, 1899, and May 31, 1902, provided the recipient had not received a clasp for a previous action in Cape Colony or the clasp NATAL.
* • NATAL: Service in Natal between October 11, 1899, and June 11, 1900.
* • RHODESIA: Service in Rhodesia between October 11, 1899, and May 17, 1900.
* • RELIEF OF MAFEKING: Troops who marched to relieve Mafeking from May 4 to 17, 1900.
* • DEFENCE OF KIMBERLEY: Troops garrisoned in Kimberley between October 14, 1899, and February 15, 1900.
* • TALANA: Troops who served north of a line drawn through Waschbank Station, October 20, 1899.
* • ELANDSLAAGTE: Troops who served at Elandslaagte on October 21, 1899.
* • DEFENCE OF LADYSMITH: Troops who defended Ladysmith between November 3, 1899, and February 28, 1900.
* • BELMONT: Troops north of Witteputs on November 23, 1899.
* • MODDER RIVER: Troops who served north of Honey Nest Kloof, November 28, 1899.
* • TUGELA HEIGHTS: Troops employed in operations north of Chieveley Station, February 14–27, 1900.
* • RELIEF OF KIMBERLEY: Troops in the relief column who marched to relieve Kimberley, February 15, 1900.
* • PAARDEBERG: Troops who served within seven thousand yards of General Cronje's laager or Koodoe's Rand Drift, February 17–26, 1900.
* • ORANGE FREE STATE: Troops in the Orange River Colony between February 28, 1900, and May 31, 1902.
* • RELIEF OF LADYSMITH: Troops who relieved Ladysmith between December 15, 1899, and February 28, 1900.
* • DRIEFONTEIN: Troops who advanced from Poplar Grove, March 10, 1900.
* • WEPENER: Troops engaged in Wepener, April 9–25, 1900.
* • DEFENCE OF MAFEKING: Troops in Mafeking between October 13, 1899, and May 17, 1900.
* • TRANSVAAL: Troops who served in the Transvaal between May 24, 1900, and May 31, 1902.
* • JOHANNESBURG: Troops who were north of Klip River Station and east of Krugersdorp Station, May 29, 1900.
* • LAING'S NEK: Troops employed in operations north of Newcastle, June 2–9, 1900.
* • DIAMOND HILL: Troops who were in an operational area east of Silverton Siding and Vlakfontein, June 11–12, 1900.
* • WITTEBERGEN: Troops who served in an operational area between Harrismith to Bethlehem and Senekal to Clocolan, July 1–29, 1900.
* • BELFAST: Troops who were east of Wonderfontein on August 26 or 27, 1900.
Naming: Impressed capitals, occasionally running script for certain British units.
Other: Members of the Lord Strathcona's Horse received approximately three hundred medals with raised dates.
Number Awarded to Canadians: 3,860.
King Edward's South Africa Medal, 1901–02 (King's South Africa Medal)
King's South Africa Medal obverse.
Origins: Queen Victoria having died in 1901, King Edward VII desired a medal bearing his effigy to be issued to those serving in the South African War.
Criteria: Service in South Africa for those who served on or after January 1, 1902, and had completed eighteen months of service before June 1, 1902. It was always issued in conjunction with the Queen's South Africa Medal.
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing an uncrowned effigy of King Edward VII in a field marshal's uniform, circumscribed by the text EDWARDVS VII REX IMPERATOR, the reverse being identical to that of the Queen's South Africa Medal. The medal was designed by George William de Saulles.
Suspender: A swivelling claw-footed suspender with a straight bar.
Ribbon: A 32 mm wide ribbon made up of equal parts green, white, and orange.
Clasps: Two clasps were issued with this medal, almost invariably in tandem:
* • SOUTH AFRICA 1901.
* • SOUTH AFRICA 1902.
Each clasp is 3.5 mm by 35 mm and has a raised edge.
King's South Africa Medal reverse.
Naming: All medals were issued named, Canadian issues in impressed capitals.
Other: This medal could only be awarded without clasps to nursing sisters.
Number Awarded to Canadians: 195.
The First World War Medals (1914–19)
Six different medals were awarded to those serving in the First World War: the 1914 Star; the 1914–15 Star; the British War Medal; the Territorial Force War Medal, 1914–18; the Mercantile Marine War Medal; and the Allied Victory Medal. With the exception of the Territorial Force War Medal, for which Canadians were ineligible, Canadians received various combinations of these war medals for their service in the Great War.
The origins of the First World War are too complex to cover in this work. Europe was divided into two opposing powers at the time: the Triple Entente (British Empire, France, and the Russian Empire) and the Central Powers (German Empire and the Austro-Hungarian Empire). As the conflict escalated, other countries became involved. The spark that set off the war occurred on June 28, 1914, with the assassination of Archduke Franz Ferdinand of Austro-Hungary by a Serbian nationalist. The Austro-Hungarian Empire demanded satisfaction from the Kingdom of Serbia, a strong ally of Russia. On July 28, 1914, the Austro-Hungarian Army invaded Serbia, and its main ally, Germany, invaded the neutral countries of Belgium and Luxembourg en route to France. Belgium's neutrality had been guaranteed by the Treaty of London in 1839, which established it as an independent kingdom, and Germany's violation of the treaty brought Britain and its empire into the war.
The war spread to every corner of the globe and claimed more than twenty million civilian and military casualties. Canada entered the war on August 4, 1914, alongside Britain and all the other dominions, colonies, and territories in the British Empire.
The 1914 Star
1914 Star obverse. (Clasp also shown.)
Origins: Created to recognize those who served in the very early stages of the First World War, notably those members of the "Old Contemptibles," Britain's national army of the period.
Criteria: Service in France or Belgium as part of the British Expeditionary Force between August 5, 1913, and November 22–23, 1914.
Insignia: A bronze four-pointed star 45 mm wide and 57 mm height, the obverse depicting a Tudor Crown where the uppermost point of the star would be, a pair of crossed swords protrude from the star, a laurelled wreath in the centre bearing a ribbon with the dates AUG — 1914 — NOV, with a modified version of the Royal Cypher GV at the base of the wreath. The reverse of this medal is plain.
Suspender: A ring integral to the insignia is struck as part of the cross atop the crown. The ribbon passes through this ring.
Ribbon: A 32 mm wide watered ribbon of equal stripes red, white, and blue.
Clasps: A bronze clasp 32 mm wide and 5 mm high with raised border bearing the text 5TH AUG–22ND NOV 1914 could be issued with this medal. However, none were awarded to Canadians.
Naming: Impressed capitals, number, rank, initials, surname, and unit.
Other: This medal was always issued with the British War Medal and Victory Medal. The medal was created in 1917, with the bar being approved on October 19, 1919. Erroneously known as the Mons Star, a total of 378,000 were issued to members of the British Expeditionary Force. Canadian recipients were all members of the 2nd Canadian Stationary Hospital.
Number Awarded to Canadians: 160.
The 1914–15 Star
1914–15 Star obverse.
1914–15 Star reverse.
Origins: Similar to the 1914 Star, the 1914–15 Star was created in 1918 to recognize those who had served in the early stages of the First World War in any theatre of war.
Criteria: Awarded to those serving the Crown in a theatre of war between August 5, 1914, and December 31, 1915. Overseas service was considered to be 4.8 kilo-metres (three-mile limit) beyond the Canadian coast for those serving in the Royal Naval Canadian Volunteer Reserve (RNCVR).
Insignia: Similar to the 1914 Star with the only difference being the inscription on the ribbon in the centre of the star that reads 1914–15.
Suspender: A ring integral to the insignia is struck as part of the cross atop the crown. The ribbon passes through this ring.
Ribbon: Identical to the 1914 Star ribbon.
Clasps: None.
Naming: Impressed capitals, number, rank, initials, surname and unit.
Other: This medal was always issued with the British War Medal and Victory Medal. A total of 2,366,000 were awarded to British and empire forces. Those who qualified for the 1914 Star could not receive this star.
Number Awarded to Canadians: 71,150.
The British War Medal (1914–20)
Origins: The British War Medal was created by King George V to "record the bringing of the war to a successful conclusion, and the arduous services rendered by His Majesty's Forces."[5]
Criteria: Awarded to all those who served overseas as members of the Canadian military forces (also St. John Ambulance and Canadian Red Cross) between August 5, 1914, and November 11, 1918, and those who served in the Canadian Siberian Expeditionary Force into June 1919. Canadian personnel who enlisted in the United Kingdom and did not serve in a theatre of war were not entitled to this medal. Those serving in the Royal Air Force were subject to the same criteria as the army, while naval personnel had to have twenty-eight days of mobilized service. Members of the Canadian Merchant Marine had to serve six months on government ships or ships of the Merchant Marine. Loss of life while in active operations before completing the period of qualification resulted in automatic qualification for this medal.
1914–18 British War Medal obverse.
1914–18 British War Medal reverse.
Insignia: A circular silver medal 36 mm in diameter bearing on the obverse an uncrowned effigy of King George V circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:. The reverse depicts an unclothed St. George holding a sword, astride a horse trampling the heraldic shield of Prussia and the skull and crossbones, with the rising sun of victory in the background along with the dates 1914–1918. This medal was designed by William McMillan, with the effigy of the King undertaken by Sir Bertram Mackennal.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A 32 mm wide watered ribbon with a central 16 mm stripe of orange, flanked on each side by 3 mm of white, 2 mm of black, and 3 mm of blue.
Clasps: No clasps were issued with this medal, though it is interesting to note that the original plan was for clasps to be issued denoting specific actions and campaigns. The British Admiralty initially published a list of forty-four clasps, but the original plan was scrapped because of its immense cost and complexity. It was projected that some recipients of the medal would be entitled to no fewer than twenty-nine clasps.
Naming: Capital letters, number, abbreviated rank, initials, surname, and unit for men; abbreviated rank, initials, and surname for officers.
Other: A total of 6,500,000 were issued throughout the British Empire.
Number Awarded to Canadians: 427,993.
The Mercantile Marine War Medal
Origins: Throughout the First World War members of the Merchant Navy and those serving on ships under the auspices of the British Board of Trade kept the supply lines between Britain, North America, and mainland Europe open in the face of the Imperial German Navy. During the war, the U-boat menace was a significant threat to supply lines and the transit of personnel.
Mercantile Marine War Medal obverse.
Mercantile Marine War Medal reverse.
Criteria: One voyage through a danger zone or six months of service at sea in the Board of Trade or Merchant Navy between August 4, 1914, and November 11, 1918. Licensed pilots, fishermen and crews of pilotage and lighthouse authorities' vessels, and post office cable ships were also included.
Insignia: A circular bronze medal 36 mm in diameter bearing on the obverse an uncrowned effigy of King George V circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:. The reverse depicts a merchant ship in rough seas with a U-boat sinking in the background and a sailing ship in the distance. The entire scene is surrounded by a laurel wreath with text on three lines: FOR WAR SERVICE MERCANTILE MARINE 1914–1918. This medal was designed by Harold Stabler, with the effigy of the King undertaken by Sir Bertram Mackennal.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A 32 mm wide watered ribbon half green, half red, with a central 4 mm stripe of white.
Clasps: None.
Naming: Impressed or engraved capitals, initials, and surname.
Other: A total of 133,135 were issued throughout the British Empire.
Number Awarded to Canadians: 624.
The Allied Victory Medal (1914–19)
Origins: The Allied Victory Medal was struck and awarded to mark the end of the First World War and victory over the Central Powers. The general concept for a medal common to all of the Allies is said to have come from France's Marshal Foch, the supreme commander of the Allied Forces.
Criteria: Service in a theatre of war as a member of the military or on the establishment of a unit between August 5, 1914, and November 11, 1918. Those who served in the Canadian Siberian Expeditionary Force were also awarded this medal. This medal was always awarded with the British War Medal, 1914–1918.
Insignia: A circular bronze medal 36 mm in diameter depicting the standing figure of the winged Victory, with her arm extended holding the branch of peace in her right hand. The reverse carries the legend THE GREAT WAR FOR CIVILISATION 1914–1919 in four lines, the entire text circumscribed by a wreath. This medal was designed by William McMillan.
Allied Victory Medal obverse.
Allied Victory Medal reverse.
Suspender: A bronze ring 13 mm in diameter.
Ribbon: A 38 mm wide watered ribbon consisting of a central 3 mm stripe of red flanked on each side by 6 mm yellow, 2 mm green, 4 mm blue, and 3 mm violet.
Clasps: No clasps were awarded with this medal, though the Mention in Dispatches oak branch device was worn on the ribbon by those entitled to it.
Naming: Capital letters, number, abbreviated rank, initials, surname, and unit for men; abbreviated rank, initials, and surname for officers.
Other: A similar design and ribbon was used by other Allies, including Belgium, Brazil, Cuba, Czechoslovakia, France, Greece, Italy, Japan, Poland, Portugal, Thailand, and the United States. The Union of South Africa issued an identical version, though the legend on the reverse was displayed in both English and Afrikaans. A total of 5,725,000 were issued throughout the British Empire.
Number Awarded to Canadians: 351,289.
The Second World War Campaign Stars and War Medals (1939–45)
Origins: A series of thirteen medals created between 1943 and 2012 recognize those who served in a myriad of operations and campaigns that took place during the largest global conflict that has ever occurred. All of the Second World War campaign stars, the Defence Medal, and the 1939–45 War Medal were awarded to servicemen and women throughout the British Empire. Along with Canada, Australia, New Zealand, South Africa, Newfoundland, India, and Rhodesia issued a dominion-specific war service medal that was often awarded in addition to the aforementioned medals.
Not surprisingly, the origins of the Second World War are found in the First World War. On September 1, 1939, Nazi Germany invaded Poland, having previously annexed Austria and parts of Czechoslovakia in 1938 and 1937 respectively. Britain declared war on Germany on September 3, and Canada followed suit seven days later: it was the first time that Canada had declared war as an independent country. Italy declared war on Britain and France on June 10, 1940, and Canada reciprocated by declaring war on Italy. Canada declared war on the Japanese Empire the day after the Japanese attack on Pearl Harbor on December 7, 1941. Canadians saw service in every theatre of operations and played an especially important role in the Battle of Britain, the Battle of the Atlantic, the defence of Hong Kong, the bombing of Germany, the invasion of Sicily and the rest of Italy, and the D-Day landings and liberation of Europe.
The creation of the Second World War campaign stars followed in the tradition of the 1914 Star and the 1914–15 Star, which were awarded for service in the early part of the First World War. The immense scope of the conflict that broke out on September 1, 1939, and did not conclude until September 2, 1945, precluded the issuing of clasps to the 1939–45 War Medal, and thus the system of campaign-specific stars with clasps was established. A maximum of five stars could be earned. Once the maximum of five stars was reached, clasps were awarded to signify service in other theatres that would have been recognized with another campaign star had the individual not already reached the maximum of five stars. Clasps were also issued to indicate participation in a particular action or event, most notably the Battle of Britain and service in the 8th Army in North Africa.
In 2012 the British government announced the creation of a ninth star in this series, the Arctic Star, which is to be awarded for service in the Arctic convoys. With the addition of this new medal, it is possible for a person to have earned six campaign stars. Simultaneous with this the British government also created a Bomber Command clasp to the 1939–1945 Star. However, Canadians are not eligible for this, since they are entitled to the Bomber Command bar that is awarded to accompany the Canadian Volunteer Service Medal.
The campaign stars were awarded to service personnel throughout the British Empire and Commonwealth. Those issued to Canadians were struck at the Royal Canadian Mint from punches provided by the Royal Mint. There is a diversity of variations since some of the stars continue to be struck as replacements. The design for all nine stars was devised by the Royal Mint design staff, while the ribbon designs were allegedly developed by King George VI.
The 1939–45 Star
1939–45 Star obverse.
Battle of Britain clasp.
Origins: To recognize services in operations rendered between September 3, 1939, and September 2, 1945.
Criteria: Granted for a period of not less than six months' service in operations from September 3, 1939, to September 2, 1945. Those evacuated from Dunkirk and Norway were also eligible for this medal, despite not having the requisite service. The Battle of Britain clasp was granted to aircrew of fighter aircraft engaged in the Battle of Britain between July 10 and October 31, 1940.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE 1939–1945 STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon of equal stripes, navy blue, red, and sky blue. A gold rosette is worn on the undress ribbon to signify receipt of the Battle of Britain clasp.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text BATTLE OF BRITAIN or BOMBER COMMAND. Canadians are ineligible for this second clasp.
Other: For many prisoners of war this would be the only campaign star they earned.
Number Awarded to Canadians: 305,000 (88 with the Battle of Britain clasp).
The Atlantic Star
Atlantic Star obverse. (France and Germany and Air Crew Europe clasps also shown.)
Origins: The Atlantic Star recognized service in the Battle of the Atlantic, primarily in recognition of convoy, escort, anti-submarine, and fast merchant ships that sailed without the benefit of escort.
Criteria: Granted for a six-month service afloat between September 3, 1939, and May 8, 1945. Time as a prisoner of war could not be counted toward this medal. The Air Crew Europe clasp was awarded when the qualifying period for that medal had been achieved after the Atlantic Star had been earned. Similarly, the France and Germany clasp was awarded when the qualifying period for that medal had been achieved after the Atlantic Star had been earned.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE ATLANTIC STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon of equal parts watered blue, white, and sea green. A silver rosette is worn on the undress ribbon to signify receipt of either of the clasps.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text AIR CREW EUROPE or FRANCE AND GERMANY.
Other: Recipients of this medal who became eligible for either the Air Crew Europe Star or France and Germany Star received the Air Crew Europe or France and Germany clasp. Only one clasp could be worn.
Number Awarded to Canadians: 43,500.
The Air Crew Europe Star
Air Crew Europe Star.
Atlantic clasp.
Origins: The Air Crew Europe Star recognized operational flying from the United Kingdom over Europe and the United Kingdom.
Criteria: Two months' service in air crew between September 3, 1939, and June 5, 1944. Those involved in operations at sea could not qualify for this medal. The Atlantic clasp was awarded when the qualifying period for that medal had been achieved after the Air Crew Europe Star had been earned. Similarly, the France and Germany clasp was awarded when the qualifying period for that medal had been achieved after the Air Crew Europe Star had been earned.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE AIR CREW EUROPE STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon of sky blue edged on each side by 2 mm of yellow and 4 mm of black, the latter being the outermost. A silver rosette is worn on the undress ribbon to signify receipt of either of the clasps.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text ATLANTIC or FRANCE AND GERMANY.
Other: Recipients of this medal who became eligible for the Atlantic or France and Germany Star received the Atlantic or France and Germany clasp. Only one clasp could be worn.
Number Awarded to Canadians: 12,800.
The Arctic Star
Arctic Star obverse.
Origins: The Arctic Star was created in 2012 after a great deal of lobbying by various veterans' groups to recognize service in the Arctic convoys to northern Russia and their escorts during the Second World War.
Criteria: Awarded to those serving the Crown for operational service of any length of time between September 3, 1939, and May 8, 1945, north of the Arctic Circle.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE ARCTIC STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central stripe of 4 mm white flanked on each side by 1 mm of black, 4 mm of red, 6 mm of navy blue, and 3 mm of light blue.
Clasps: None.
Other: With the addition of this star it is possible for a person to have earned six campaign stars.
Number: Ongoing issue.
The Africa Star
Africa Star obverse. (1st Army, 8th Army, and North Africa 1942–43 clasps also shown.)
Origins: The Africa Star was awarded to recognize operational service in North Africa.
Criteria: Operational service in North Africa between June 10, 1940, and May 12, 1943. This includes operations in Abyssinia, Somaliland, Eritrea, and Malta. Service in West Africa did not count toward this medal. The 8th Army and 1st Army clasps were awarded for service in the 8th and 1st Armies between October 23, 1942, including the Battle of El Alamein, up to May 12, 1943. The North Africa 1942–43 clasp was awarded for service with the 18th Army Group HQ and naval or air forces between October 23, 1932, and May 12, 1943.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE AFRICA STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central 10 mm stripe of red bordered by 4 mm of sand brown, 2 mm of navy blue on the left, a 2 mm stripe of light blue on the right side, and edged in sand brown. A silver rosette is worn on the undress ribbon to signify the North Africa 1942–43 clasp, a numeral 8 for the 8th Army clasp, and a numeral 1 for the 1st Army clasp.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text NORTH AFRICA 1942–43 or 8TH ARMY or 1ST ARMY.
Other: Nil.
Number Awarded to Canadians: 7,400.
The Pacific Star
Origins: The Pacific Star was awarded to recognize service in the Pacific theatre.
Criteria: Operational service in the Pacific theatre between December 8, 1941, and September 2, 1945. For the army, operational service in China and Malaya from December 8, 1941, to February 15, 1942, also qualifies for this medal, while service from February 16, 1942, to September 2, 1945, qualifies for the Burma Star. The Burma clasp was awarded when the qualifying period for that medal had been achieved after the Pacific Star had been earned.
Pacific Star obverse. (Burma clasp also shown.)
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE PACIFIC STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central 3 mm yellow stripe, with 6 mm of green on each side, 3 mm of navy blue on the left, and 3 mm of light blue on the right, each side edged in 6 mm of red. A silver rosette is worn on the undress ribbon to signify receipt of the Burma clasp.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text BURMA.
Other: Recipients of this medal who became eligible for the Burma Star received the Burma clasp.
Number Awarded to Canadians: 8,800 (900 with the Burma clasp).
The Burma Star
Burma Star obverse. (Pacific clasp also shown.)
Origins: The Burma Star was awarded to recognize service in the Burma campaign.
Criteria: Operational service in the Burma campaign between December 11, 1941, and September 2, 1945. Operational service on land in the Indian provinces of Bengal and Assam between May 1, 1942, and December 31, 1943, and operational service on land in Bengal and Assam east of Brahmaputra between January 1, 1944, and September 2, 1945, also qualifies for this medal. The Pacific clasp is awarded when the qualifying period for that medal has been achieved after the Burma Star has been earned.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE BURMA STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central 12 mm stripe of red flanked on each side by 3 mm of navy blue, 3 mm of orange, and 3 mm of navy blue. A silver rosette is worn on the undress ribbon to signify receipt of the Pacific clasp.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text PACIFIC.
Other: Recipients of this medal who became eligible for the Pacific Star were awarded the Pacific clasp.
Number Awarded to Canadians: 5,500 (1,000 with Pacific clasp).
The Italy Star
Italy Star obverse.
Origins: The Italy Star was awarded to recognize service in the Italian campaign.
Criteria: Operational service in Sicily or in Italy between June 11, 1943, and May 8, 1945. Operational service in the Mediterranean theatre (Aegean, Dodecanese, Corsica, Greece, Sardinia, and Yugoslavia) after June 11, 1943, is a qualification. Operational service in Elba is a qualification. Service is not a qualification in Sicily after August 17, 1943; in Sardinia after September 19, 1943; and in Corsica after October 4, 1943.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE ITALY STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central stripe of green flanked on each side with equal stripes of white and red.
Clasps: None.
Number Awarded to Canadians: 91,000.
The France and Germany Star
France and Germany Star obverse.
Origins: The France and Germany Star was awarded to recognize operational service in France, Belgium, Holland, or Germany.
Criteria: Operational service in France, Belgium, Holland, or Germany between June 6, 1944, and May 8, 1945. The Atlantic clasp or the Air Crew Europe clasp are awarded when the qualifying period for that medal has been achieved after the France and Germany Star has been earned.
Insignia: A six-pointed bronze star 44 mm wide and 38 mm wide displaying in the centre the Royal Cypher GRI VI surmounted by a Tudor Crown and circumscribed by a ribbon with the text THE FRANCE AND GERMANY STAR.
Suspender: A 6 mm bronze ring.
Ribbon: A 32 mm wide ribbon with a central stripe of red flanked on each side with equal stripes of white and blue. A silver rosette is worn on the undress ribbon to signify receipt of the Atlantic clasp.
Clasps: A bronze clasp 32 mm by 5 mm with a scalloped edge and one hole in each corner displaying the text ATLANTIC.
Number Awarded to Canadians: 230,000.
The Defence Medal (1939–45)
Defence Medal obverse.
Defence Medal reverse.
Origins: The Defence Medal was awarded to commemorate certain non-operational and civil defence services rendered during the Second World War.
Criteria: Service between September 2, 1939, and September 2, 1945, consisting of six months of service in non-operational areas outside Canada that were subject to air attack or closely threatened; one year of non-operational service in His Majesty's forces outside Canada; six months in approved Civil Defence Services in non-operational areas subject to attack or closely threatened; or three years for Civil Defence or other specified civilian service in military operational areas, provided the civil category was not eligible for campaign stars.
Insignia: A silver medal 36 mm in diameter on the obverse depicting King George VI uncrowned circumscribed by his titles GEORGIVS VI D:G:BR:OMN: REX F:D: IND:IMP. The reverse depicts a Royal Crown resting atop the stump of an oak tree and flanked on each side by a lion and lioness. The dates 1939 and 1945 rest just above each lion respectively, while the base of the medal carries the text THE DEFENCE MEDAL. Designed by H.W. Parker, this medal was struck by the Royal Canadian Mint in .800 silver.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A green ribbon 32 mm wide with a central 12 mm stripe of orange bordered on each side by 4 mm of green, 2 mm of black, and 4 mm of green.
Clasps: None.
Other: Service in Newfoundland by Canadians and service in Hong Kong also counted toward the Defence Medal. Bomb and mine clearance personnel had a reduced qualification period of three months. The Commonwealth issue of this medal was struck in cupronickel.
Number Awarded to Canadians: 325,000.
The Canadian Volunteer Service Medal (1939–47)
Origins: The Canadian Volunteer Service Medal was the first medal in the Commonwealth to recognize voluntary service in the Second World War. Australia, New Zealand, India, South Africa, and Rhodesia all followed suit with similar medals. This medal is curious in that it was created and announced in 1943 in the middle of the war. It was in some ways a public relations effort on the part of the government, which was soon to enter a conscription crisis and anxious to demonstrate its recognition for those who had voluntarily enlisted to serve King and country. In addition to the clasp for overseas service, clasps have been created for Dieppe (1994), Hong Kong (1995), and Bomber Command (2013).
Canadian Volunteer Service Medal with overseas clasp obverse.
Canadian Volunteer Service Medal reverse.
The design of the obverse was arrived at through a competition open to members of the Royal Canadian Navy, Canadian Army, and Royal Canadian Air Force, with the prize set at $300 in Victory Bonds. There is no record of the winner; however, the formal artwork was undertaken by Charles Comfort, then a major in the army and a war artist. He had seven members of the military march up and down a parade square where they were photographed. From the photographs he created the design. The figures depicted on the medal were actual members of the Canadian military: Leading Seaman P.G. Colbeck (Royal Canadian Navy), WREN P. Mathie (Women's Royal Canadian Naval Service), Private D.E. Dolan (1st Canadian Parachute Battalion), Lance Corporal J.M. Dann (Canadian Women's Army Corps), Flight Sergeant K.M. Morgan (Royal Canadian Air Force), Leading Air Woman O.M. Salmon (RCAF[Women's Division]), and Lieutenant Nursing Sister E.M. Lowe (Royal Canadian Army Medical Corps).
Criteria: Eighteen months of voluntary service in the military forces of Canada, Merchant Navy, members of the Corps of Canadian (Civilian) Fire Fighters, Canadian Red Cross, St. John Ambulance, Ferry Command, YMCA, Salvation Army, Canadian Legion, British Commonwealth Air Training Plan Instructors, or Knights of Columbus between September 3, 1939, and March 1, 1947. When first established on October 22, 1943, only members of the military forces of Canada and Merchant Navy were eligible for the medal. The medal was automatically awarded to those who voluntarily served and were killed in action or died as a result of an injury or disease attributable to active service.
Insignia: A circular silver medal 36 mm in diameter, the obverse depicting seven marching servicemen and women, representing the Royal Canadian Navy, Canadian Army, Royal Canadian Air Force, and the nursing service. The vignette is circumscribed by the text 1939 CANADA 1945 VOLUNTARY SERVICE VOLONTAIRE, with small natural maple leaves on each side of the word SERVICE. The reverse of the medal depicts the 1921 version of the Royal Arms of Canada. This medal was designed by Charles Comfort and Thomas Shingles and was struck by the Royal Canadian Mint in .925 silver.
Suspender: A straight graduated silver suspender 38 mm in width at its widest point attached to the top of the disc via a link and eyelet that is soldered on the top of the disc.
Ribbon: A 32 mm wide ribbon with a centre 13 mm wide in royal blue flanked on each side by equal stripes of red and dark green at the outermost edge.
Clasps: Four different clasps were issued with this medal:
Hong Kong clasp.
Dieppe clasp.
Bomber Command clasp.
* • OVERSEAS SERVICE: A silver clasp 33 mm wide and 5 mm high bearing a natural maple leaf in the centre. Awarded for sixty days of service outside Canada.
* • DIEPPE: A silver clasp 33 mm wide and 4 mm high with a frosted background bearing the text DIEPPE, the bar surmounted by an anchor, rifle, and eagle. Awarded to those who participated in the Battle of Dieppe on August 19, 1942.
* • HONG KONG: A silver clasp 33 mm wide and 5 mm high with a raised edge and the text HONG KONG, with the HK service patch between the two words. Awarded to those who participated in the Battle of Hong Kong, December 8–25, 1941.
* • BOMBER COMMAND: A silver clasp 33 mm wide and 5 mm high with a polished edge and frosted background displaying in the centre a Second World War heavy bomber seen from above. Awarded to flying and non-flying personnel for operational service with Bomber Command. The clasp was designed by Deputy Chief Herald Bruce Patterson.
Other: This medal could be awarded without a clasp; there was no maximum to the number of clasps a recipient could receive. Later issues struck by the Royal Canadian Mint are found with a matte or brushed finish and occasionally a slightly raised edge.
Number Awarded to Canadians: 650,000 with 525,000 overseas clasps.
The Newfoundland Volunteer Service Medal (1939–45)
Newfoundland Volunteer Service Medal obverse.
Newfoundland Volunteer Service Medal reverse.
Origins: When the Second World War broke out, Newfoundland was a dominion and not yet a Canadian province. Thus, Newfoundlanders who served in any of the British forces, as well as those who served in the Merchant Navy and the Royal Newfoundland Regiment, were unrecognized for their voluntary service. In 1981, to rectify this problem, the Government of Newfoundland sanctioned the striking of a bronze medal that is in essence a Newfoundland version of the Canadian Volunteer Service Medal.
Criteria: Awarded for service between September 3, 1939, and September 2, 1945, to any person who volunteered and served in units or organizations raised or maintained as the contribution of the Dominion of Newfoundland as part of the war effort. Notably, service with any element of the British Imperial Forces, the Newfoundland Overseas Forestry Unit, the Merchant Navy, and the Royal Newfoundland Regiment that has not been recognized by a volunteer war service medal from another country is eligible (i.e., receipt of the Canadian Volunteer Service Medal makes an individual ineligible).
Insignia: A bronze medal 36 mm in diameter, the obverse displaying a caribou atop the Royal Cypher of King George VI (post–India independence version GVIR) and circumscribed by the text NEWFOUNDLAND VOLUNTEER SERVICE MEDAL 1939–1945. The reverse bears a representation of Britannia in the centre flanked by two lions, with Britannia protected by a scallop shell. The medal was designed and struck by the Royal Mint.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A maroon ribbon 36 mm wide edged on each side by 1.5mm of blue, 1.5 mm of white, and 1.5 mm of red.
Clasps: None.
Naming: Each medal is impressed with an issue number.
Other: Presented in a red leatherette box from the Royal Mint.
Number: 4,630.
The 1939–45 War Medal
1939–45 War Medal obverse.
1939–45 War Medal reverse.
Origins: The 1939–45 War Medal was awarded to recognize service rendered in any theatre of the Second World War.
Criteria: Operational or non-operational service between September 3, 1939, and September 2, 1945, of at least twenty-eight days, including service in the Merchant Navy.
Insignia: A circular silver medal 36 mm in diameter, the obverse bearing a crowned effigy of King George VI circumscribed by the legend GEORGIVS VI: D: BR: G: OMN: REX ET INDIAE IMP:. The reverse depicts a lion standing triumphant over the body of a double-headed dragon, the dates 1939 and 1945 in the upper exergue. The medal was designed by Edward Carter-Preston, and the Canadian issue was struck at the Royal Canadian Mint in .800 silver.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A ribbon 32 mm wide with a central 2 mm stripe of red bordered on each side by 2 mm of white, 5 mm of dark blue, and 5 mm of red.
Clasps: No clasps were awarded with this medal, though the Mention in Dispatches oak leaf device was worn on the ribbon by those entitled to it.
Naming: Only those awarded to members of the Canadian Merchant Navy were named. Initials and surname are in engraved capitals (approximately 4,500).
Other: The Commonwealth issue of this medal was struck in cupronickel.
Number Awarded to Canadians: 700,000.
The Queen's Korea War Medal (1950–53)
Origins: After the end of the Second World War, Korea was divided into two parts along the 38th parallel: the northern part, the Democratic People's Republic, was within the sphere of influence of Communist China and the Soviet Union; the southern part, the Republic of Korea, was under the protection of the Western powers. Almost from the outset, tensions rose between the north and south as the Cold War escalated. On June 25, 1950, North Korean forces crossed the 38th parallel and invaded South Korea. This brought condemnation from the newly formed United Nations and the creation of an international U.N. force involving the United States, United Kingdom, Australia, Belgium, Canada, Colombia, Ethiopia, France, Greece, Luxembourg, Netherlands, New Zealand, Philippines, South Africa, Thailand, and Turkey. The war lasted until July 27, 1953, when an armistice was achieved. More than twenty-six thousand Canadians participated in the war, with 516 deaths — 316 of which were as a result of combat.
As with the Second World War, Canada decided to issue its own war medal for this conflict. The medal was originally supposed to carry the effigy of King George VI who approved the Commonwealth version. However, his death in 1952 saw this plan abandoned. While the Canadian medal, approved later by the new Queen, shares the same ribbon and is similar to the Commonwealth design, the Canadian medal was struck by the Royal Canadian Mint in .800 silver and has a slightly different obverse, making it distinctly Canadian. The Canadian Cabinet originally gave consideration to adding a CANADA bar to the medal to enhance its "Canadianness." However, the simpler solution of adding the word to the obverse of the medal was opted for as part of the Queen's Canadian title.
Queen's Korea War Medal obverse.
The Commonwealth version of the medal was struck in cupronickel and awarded to members of the British forces and those of Australia and New Zealand. South Africa, which also made a modest contribution to the Korean War, struck its own unique medal.
Criteria: One day of service in Korea on the strength of a Canadian Army unit, twenty-eight days afloat within the waters of the war zone, or one sortie by a member of the Royal Canadian Air Force between July 2, 1950, and July 27, 1953.
Insignia: A silver medal 32 mm in diameter, the obverse bearing an uncrowned effigy of Queen Elizabeth II and the text ELIZABETH II DEI GRATIA REGINA with CANADA, the obverse depicting the Lernaean Hydra (multi-headed snake) being slain by Hercules with the text KOREA at the base. While the Canadian issue was struck in .800 silver, the Commonwealth issue was struck in cupronickel and carried a slightly different set of titles for the Queen on the obverse. The obverse effigy was sculpted by Mary Gillick; the reverse was designed by Edward Carter-Preston.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: A 32 mm wide ribbon composed of alternating equal stripes, three in yellow and two in blue.
Clasps: No clasps were awarded with this medal, though the Mention in Dispatches oak leaf device was worn on the ribbon by those entitled to it.
Queen's Korea War Medal reverse.
Naming: Engraved or impressed capitals, number, initials, and surname.
Other: This medal was always issued with the U.N. Medal for Korea, and all those eligible for these two medals were also eligible for the Canadian Volunteer Service Medal that the Government of Canada created much later (see chapter 23). The striking of this medal began the tradition of including the text CANADA along with the Queen's title in Latin. This practice has been maintained on all but three of the Canadian medals issued bearing an effigy of the Sovereign to date.
Commonwealth issue Queen's Korea War Medal obverse.
Number: 15,000.
# 7
IMPERIAL LONG SERVICE DECORATIONS AND MEDALS
There is perhaps no more confusing element of the imperial honours used in Canada than the various long service decorations and medals. The myriad of long service medals and the varying awarding criteria is complex and occasionally misunderstood. When one considers the fact that there were individual dominion/colonial issues of many of these medals, such as the Efficiency Medal, with more than forty-four different suspension bars for places such as Gibraltar, Antigua, Canada, India, Ceylon, and many others, the scope of the system is fully revealed.
Royal Navy Long Service and Good Conduct Medal George IV obverse.
Royal Navy Long Service and Good Conduct Medal George IV reverse.
The evolution of imperial long service awards in Canada provides an interesting window into Canada's role in the broader British Empire. All of the imperial long service awards covered in this chapter were also, in one form or another, awarded to servicemen in other parts of the empire. The Navy Long Service and Good Conduct Medal awarded to the British, Australians, and New Zealanders was identical, though it would have included RN, RAN, or RNZN impressed on the rim of the medal as opposed to RCN.
Until the creation of the Canadian Forces' Decoration, the British approach to long service recognition prevailed in Canada. Officers serving as regular members of the Royal Navy and British Army were not entitled to receive any long service decorations or medals, since being a commissioned officer and servant of the Crown was believed to be sufficient recognition. Regular non-commissioned officers in the navy, army, and air force were entitled to circular long service medals. Officers in the reserve (territorial) army, the Royal Navy Reserve, and the Royal Navy Volunteer Reserve were awarded decorations, and these were invariably oval in shape and entitled the recipient to use postnominals. Privates, non-commissioned officers, ratings, warrant officers, and petty officers in the reserve components of the army, navy, and air force were entitled to various long service medals, none of which entitled the recipient to postnominals. The entire system was based on branch of service, rank, and whether or not you were a part-time or full-time member.
These awards form the foundation for the modern Canadian Forces' Decoration, and the complexity of this system of "service- and rank-specific awards" in which one received a certain medal for serving in the navy, army, or air force is still the norm for most Commonwealth countries. Since 1999 the British moved to a more limited number of long service awards that are not necessarily specific to service in a particular branch of their military.
All of these medals were struck in silver (.800 or .925) and named. The only award covered in this chapter that was issued with a certificate was the Imperial Service Medal.
It is interesting to note that when the British government set about creating the medals covered in this chapter, they never consulted with Canadian or other Commonwealth governments. They simply created the awards and notified the dominions about the new medals through a memo from the Colonial/Dominions Office. By all accounts, Canada was happy to participate and award the various long service medals because they were not only marks of distinction, but it was far cheaper for the Department of Militia and Defence to use a pre-existing British award than to go to the trouble of creating a specific Canadian award for what was a fairly small military.
Not covered in this chapter are the various colonial police long service medals that colonies and dominions such as Ceylon (Sri Lanka), South Africa, and Cyprus created to recognize long service in their respective constabularies. Aside from the Royal Canadian Mounted Police, Canada never created any dominion-specific long service award for its police forces. This was largely on account of the fact that policing is largely a provincial (or, by extension, municipal) responsibility and not centralized, except for the RCMP. Newfoundland had also briefly experimented with a long service medal for the Royal Newfoundland Constabulary. Unlike other dominions and colonies that had highly centralized police services, Canada had no central organizing authority to administer such awards as the Colonial Police Long Service Medal; thus, they were never awarded in Canada.
There are a wide variety of naming styles for these medals. Some were engraved in the United Kingdom at the Royal Mint, while others were engraved in Canada by either the Royal Canadian Mint or a number of private firms. Medals were always named and they were not issued unnamed. This chapter begins with the Meritorious Service Medal, a unique form of long service recognition, and is followed by the other naval, military, and air force long service honours grouped by the date in which a service was established in Canada (army, navy, air force), then on the basis of regular/permanent or reserve service. The chapter closes with the Imperial Service Medal, which was awarded to civilians.
The Meritorious Service Medal
Meritorious Service Medal George V 1st type obverse.
Origins: The Meritorious Service Medal (MSM) was created on December 19, 1845, and is the most unusual award for long service, since it included an annuity paid to the recipient. It was usually awarded for meritorious service, though during the First World War there were a large number awarded for gallantry.
Criteria: The medal was reserved for warrant officers and senior non-commissioned officers who were discharged after more than twenty-one years of service. In the event of a disability, the period of service only had to be eighteen years. The recipient also had to have earned a Long Service Medal (either the Army Long Service and Good Conduct Medal or the Permanent Forces of the Empire Beyond the Seas Long Service and Good Conduct Medal). In Canada the number of recipients at any one time was limited to forty, which was increased to seventy-five in 1928 and one hundred in 1939. This number did not include those who were awarded the MSM during the First World War for gallantry. The gallantry provision for award of the medal was cancelled on September 7, 1928, since the British Empire Medal, which was created on December 29, 1922, fulfilled this purpose and was also a much higher-ranked award.
From 1916 until 1928, the MSM was awarded to all non-commissioned officers "duly recommended for the grant in respect of gallant conduct in the performance of military duty otherwise than an action against the enemy, or in the saving, or attempting to save the life of an officer or soldier, or for devotion to duty in a theatre of war." Awarded to Canadians from1902 to 1958, recipients of the last MSMs had to have enlisted prior to September 1, 1939, and the final Canadian recipients were Sergeant Major A.W. Grint (Royal Canadian Army Medical Corps) and Sergeant G.N. Gray (Royal Canadian Regiment).
Insignia: A circular silver medal 36 mm in diameter. There are six different obverse designs:
* • Edward VII in field marshal's uniform with the text EDWARDVS VII REX IMPERATOR.
* • George V 1st type in field marshal's uniform with the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:.
* • George V 2nd type, bareheaded, circumscribed by the text GEORGIVS V BRITT: OMN: REX: ET IND: IMP.
* • George VI 1st type, bareheaded, with the text GEORGIVS VI D:G: BR:OMN:REX F:D :IND: IMP.
* • George VI 2nd type, bareheaded, with the text GEORGIVS VI D:G: BRITT: OMN: REX FID: DEF:.
* • Elizabeth II effigy with the text ELIZABETH II DEI GRATIA REGINA: F:D.
Meritorious Service Medal reverse.
There are two reverse types for Canada:
* • The 1st type bears two sprigs of laurel leaves with a Royal Crown at the top and the text FOR MERITORIOUS SERVICE.
* • The 2nd type is identical to the first, but it carries the text CANADA at the top. This 2nd type is by far the rarest.
Suspender: A claw-footed suspender with an ornate scroll bar. Both swivelling and non-swivelling examples exist.
Ribbon: The medal was issued with three different ribbons. From 1845 to 1916, it was a 32 mm wide crimson ribbon; from 1916 to 1917, the medal had a crimson ribbon 32 mm wide edged with 2 mm of white; since 1917, the medal has had a crimson ribbon 32 mm wide edged with 2 mm of white and a central white stripe of 2 mm.
Bars: None.
Naming: Impressed or engraved capitals, abbreviated rank, initials, surname, and unit.
Postnominals: None.
Other: The MSM continues to be awarded to non-commissioned members of the British Armed Forces.
Number Awarded to Canadians: 1,705 (50 Edward VII, 60 King George V 1st type, 25 King George V 2nd type, 100 George VI both types, 40 Elizabeth II both types, 2 George V for gallantry in the RCN, 1 George V for gallantry in the RCAF, 1,427 George V for gallantry in the Canadian Army).
The Army Long Service and Good Conduct Medal
Army Long Service and Good Conduct Medal obverse.
Army Long Service and Good Conduct Medal reverse.
Origins: Although this medal was created in 1830 by King William IV, it was awarded for service in the British Army. In 1895 it was decided that the scope of this medal should be broadened to include members of the various permanent forces in the dominions and India. The obverse and reverse of medals for every dominion were the same, though the reverse of the medal included the name of the specific dominion in question. Thus, Canadian issues bore the text CANADA and Australian issues AUSTRALIA. This version of the medal was created by Royal Warrant on May 31, 1895. Despite this early creation date, it would appear that this medal was only awarded to Canadians between 1902 and 1909, after which it was superseded by the Colonial Permanent Forces Long Service and Good Conduct Medal.
Criteria: Awarded to warrant officers, non-commissioned officers, and men serving in the permanent forces for eighteen years.
Insignia: A 36 mm circular silver medal. The obverse bears an effigy of King Edward VII in a field marshal's uniform with the legend EDWARDVS VII REX IMPERATOR. The reverse bears an inscription on four lines: FOR LONG SERVICE AND GOOD CONDUCT, with CANADA at the very top slightly curved.
Suspender: A swiveling claw-footed suspender with an ornate scroll bar.
Ribbon: A crimson red ribbon 32 mm wide.
Bars: None.
Postnominals: None.
Naming: Impressed capitals, abbreviated rank, initials, surname, and abbreviated unit.
Other: A number of unnamed examples have found their way onto the market. All official issues were issued named.
Number: Although the Royal Mint struck a batch of this medal bearing Queen Victoria on the obverse and CANADA on the reverse, it appears as though all 150 of those actually awarded were King Edward VII issues.
The Permanent Overseas Forces Long Service and Good Conduct Medal
Permanent Overseas Forces Long Service and Good Conduct Medal reverse.
(1st type ribbon shown.)
Permanent Overseas Forces Long Service and Good Conduct Medal George V obverse.
(2nd type ribbon shown.)
Origins: Created by Royal Warrant in 1909, this medal was instituted to recognize long service and good conduct in the wide variety of permanent forces of British colonies and dominions. Most commonly known as the Permanent Forces of the Empire Beyond the Seas Long Service and Good Conduct Medal, it was only awarded to those serving in the various permanent forces of dominions and colonies and not to those serving in the United Kingdom. This medal was awarded to Canadians until the end of 1931 when it was replaced by the Canadian Army Long Service and Good Conduct Medal.
Criteria: Awarded for eighteen years of service in the Canadian Army, Royal Canadian Air Force, and Royal Canadian Navy (until 1925).
Insignia: A circular silver medal 36 mm in diameter. The obverse bears an effigy of King George V in field marshal's uniform circumscribed by the legend GEORGIVS V BRITT:OMN:REX ET IND:IMP:.
Suspender: A non-swivelling claw-footed suspender with a straight bar. Some of the early issues were swivel suspensions.
Ribbon: From 1909 to 1916, the ribbon was 32 mm wide and crimson red with a 2 mm central stripe of white. In 1916 the ribbon was changed to a crimson 32 mm wide ribbon with a central stripe of 3 mm dark blue bordered on each side by 3 mm of white.
Bars: None.
Postnominals: None.
Naming: Impressed capitals, abbreviated rank, initials, surname, and abbreviated unit. A few late issues are found with engraved capitals.
Other: Awarded in a small white cardboard box.
Number Awarded to Canadians: 840 (839 to the Canadian Army, 1 to the RCAF).
The Colonial Auxiliary Forces Officers' Decoration
Colonial Auxiliary Forces Officers' Decoration George V obverse.
Origins: Initially established as the Volunteer Officers' Decoration by Queen Victoria on July 25, 1892, the Volunteer Decoration (VD) was initially intended to "reward long and meritorious services of Officers of proved capacity in Our Volunteer Forces in Great Britain." The award of the VD was extended to officers of the various dominion and colonial volunteer forces on May 24, 1894. The regulations were further amended on May 18, 1899, to create the Colonial Auxiliary Forces Officers' Decoration. A similar Indian Auxiliary Forces Officers' Decoration was created at this time for award to officers in the Indian military forces.
Criteria: Awarded to officers for twenty years of long service and good conduct in the Canadian Militia. This medal was superseded by the Efficiency Decoration on December 31, 1931.
Insignia: An oval medal surmounted by a crown, 65 mm tall and 35 mm wide. Three different obverses exist for this decoration, and each is circumscribed with the text COLONIAL AUXILIARY FORCES, bearing in the centre in openwork the Royal Cypher:
* • Victoria: VRI.
* • Edward VII: ERI VII.
* • George V: GRI.
The outer ring is silver in colour, while the Royal Cypher and Royal Crown are gilt. Occasionally, the gilding is worn off because of excessive polishing. The reverse is plain. Some issues, primarily those struck by the Royal Mint, do not bear hallmarks.
Suspender: A wire connected to the medal with a small loop.
Ribbon: A 30 mm wide ribbon solid green in colour.
Colonial Auxiliary Forces Officers' Decoration Victoria obverse.
Colonial Auxiliary Forces Officers' Decoration Victoria reverse.
Bars: None.
Postnominals: VD. There are numerous stories of a comical nature related to the fact that VD is also the well-known abbreviation for the term venereal disease (now more commonly referred to as sexually transmitted infection).
Naming: The VD awarded to Canadians was invariably named. There were two naming styles: engraved and impressed. These are found in solid upper case as well as both upper and lower case. Most often the rank, name, and regiment/unit of the recipient is included. It was left up to Canadian authorities at the Department of Militia and Defence, and sometimes at the unit level, to arrange for insignia to be named. This accounts for the various naming styles.
Other: It is not unusual to find this decoration along with the Colonial Auxiliary Forces Long Service Medal, though militia regulations prohibited a recipient of the two awards from wearing them concurrently. This decoration was manufactured by a variety of firms, including Garrard & Co. and the Royal Mint.
Number: 2,760 (850 Victoria, 250 Edward VII, 1,660 George V).
The Colonial Auxiliary Forces Long Service Medal
Colonial Auxiliary Forces Long Service Medal George V obverse.
Origins: Initially created in the United Kingdom as the Volunteer Long Service and Good Conduct Medal on June 13, 1894, the award of this medal was extended to the dominions and colonies by Royal Warrant on May 18, 1899. The award for the dominions and colonies was renamed the Colonial Auxiliary Forces Long Service Medal.
Criteria: Awarded to members of the non-permanent active militia (RCNVR and RCAF Auxiliary were briefly included in this) for twenty years of accumulated service. War service counted as double.
Insignia: A circular silver medal 36 mm in diameter. The obverse is one of three effigies:
* • Victoria (issued 1896–1902): A veiled effigy of Queen Victoria circumscribed by the legend VICTORIA DEI GRATIA REGINA ET IMPERATRIX.
* • Edward VII (issued 1902–11): King Edward VII in a field marshal's uniform circumscribed by the legend EDWARUS VII REX IMPERATOR.
* • George V (issued 1911–31): King George V in a field marshal's uniform circumscribed by the legend GEORGIVS V BRITT:OMN:REX ET IND:IMP:.
Colonial Auxiliary Forces Long Service Medal reverse.
The reverse displays a shield bearing the following text surmounted by a crown: FOR LONG SERVICE IN THE COLONIAL AUXILIARY FORCES.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Ribbon: Solid green 32 mm wide.
Bars: None.
Postnominals: None.
Naming: Impressed upper case capitals, engraved capitals, and engraved upper and lower case letters.
Other: This medal is often found with the Volunteer Officers' Decoration.
Number Awarded to Canadians: 6,400 (1,350 Victoria, 750 Edward VII, 4,300 George V).
The Canadian Army Long Service and Good Conduct Medal
Canadian Army Long Service and Good Conduct Medal George VI 2nd type obverse.
Canadian Army Long Service and Good Conduct Medal Elizabeth II 1st type obverse.
Origins: This medal replaces the original Army Long Service and Good Conduct Medal and was awarded to Canadians from 1902 to 1909 when it was replaced by the Permanent Overseas Forces Long Service and Good Conduct Medal, which was awarded from 1909 to 1930. The establishment of "The Medal for Long Service and Good Conduct (Military)" on September 12, 1930, was part of the broader attempt in the United Kingdom to modernize the various army long service awards. In Canada, by army order, this medal was known as the Canadian Army Long Service and Good Conduct Medal.
Criteria: Awarded to warrant officers, non-commissioned officers, and men of the regular Canadian Army who completed eighteen years of service and whose character and conduct was irreproachable. Bars were awarded for eighteen years of additional service. Until August 1944 this medal was also awarded to permanent members of the Royal Canadian Air Force.
Canadian Army Long Service and Good Conduct Medal Elizabeth II type "Canadian" obverse.
Canadian Army Long Service and Good Conduct Medal reverse.
Insignia: A circular silver medal 36 mm in diameter. There are five different obverses:
* • George V: Crowned wearing coronation robes circumscribed by the text GEORGIVS — D — G — BRITT — OMN — REX — ET — INDIAE — IMP —.
* • George VI 1st type: Crowned and circumscribed by the text GEORGIVS VI D:G: BR: OMN: REX ET INDIAE IMP:.
* • King George VI 2nd type: Crowned and circumscribed by the text GEORGIVS VI DEI GRA: BRITT OMN: REX FID: DEF:+.
* • Queen Elizabeth II 1st type: Wearing a Tudor Crown circumscribed by the text ELIZABETH II DEI GRATIA REGINA F.D..
* • Queen Elizabeth II 2nd type: Wearing a St. Edward's Crown circumscribed by the text ELIZABETH II DEI GRATIA REGINA. This obverse was only used in Canada.
The reverse of all issues bears the text FOR LONG SERVICE AND GOOD CONDUCT.
Suspender: An inverted fleur-de-lys holding a plaque bearing CANADA.
Ribbon: A 36 mm wide ribbon maroon in colour and edged on each side by 2 mm of white.
Bars: A bar defaced with an Imperial Crown in the centre.
Naming: Impressed or engraved capitals, abbreviated rank, initials, surname, and abbreviated unit.
Postnominals: None.
Number Awarded to Canadians: 2,350 (150 Edward VII, 700 George V, 1,250 George VI, 250 Elizabeth II) along with 32 first bars, 2 second bars.
The Efficiency Decoration
Origins: The successor award to the Volunteer Decoration, the Efficiency Decoration was established by Royal Warrant on September 23, 1930, and came into effect on January 1, 1931. The Efficiency Decoration (ED) was awarded throughout the Commonwealth and British Empire. A separate top suspender bar was made for each dominion and colony, denoting the name of the country in question. Officers in the Canadian Army and Royal Canadian Air Force were eligible for this decoration, though awards to the RCAF are quite scarce.
Efficiency Decoration George V obverse.
Efficiency Decoration George VI 2nd type obverse.
Criteria: Twenty years of commissioned service on the active list. War service counted as double, and if an officer had served in the ranks, he was permitted to carry forward half of that service to count toward his ED. The twenty years of service did not have to be continuous.
Insignia: A silver oval medal 54 mm tall and 37 mm wide. Four different obverses were issued to Canadians, each consisting of the same design of a modern oval wreath with the Royal Cypher in openwork in the centre surmounted by a Royal Crown. For the issues made during the reigns of George V and George VI, the crown is a Tudor Crown, while that used on the Elizabeth II issues is a St. Edward's Crown:
Efficiency Decoration Elizabeth II obverse.
Efficiency Decoration reverse.
* • George V (1931–37): GVR.
* • George VI 1st type (1937–48): GRI.
* • George VI 2nd type (1948–53): GVIR.
* • Elizabeth II: EIIR.
The wreath is silver while the crown, cipher, and the ribbons tying the wreath are gilt. The reverse is plain and entirely silver in colour.
Suspender: A small loop is soldered onto the back of the crown. Through this passes a larger diameter ring and the ribbon. A top mounting bar is also always attached to the top of the ribbon. For Canadians this bar carried the text CANADA.
Ribbon: A 36 mm wide green ribbon bisected in the centre by 4 mm of yellow.
Bars: A bar defaced in the centre by a crown was awarded for six years' additional service.
Naming: The ED was issued with a variety of naming styles. The most common of these was engraved capitals on the reverse, just below the crown. They are also found with impressed naming (including postnominals such as DSO and MC). Both of these engraving types are found in the form of straight lines or circumscribing the top curve of the oval medal. In a very few instances it is known that late issues of the ED were issued with the naming engraved on the rim!
Postnominals: ED. Recipients who served in the British Territorial Army referred to this as the Territorial Decoration, and the postnominals TD were used in the United Kingdom.
Other: All EDs were awarded in a black leatherette case from the Royal Mint.[1]
Number Awarded to Canadians: 3,700.
The Efficiency Medal
Efficiency Medal Elizabeth II 1st type obverse.
Efficiency Medal George VI 2nd type obverse.
Origins: Established at the same time as the Efficiency Decoration, the Efficiency Medal (EM) was created to replace the Colonial Auxiliary Long Service and Good Conduct Medal that had been awarded to Canadians from 1899 to 1931. The Efficiency Medal was awarded primarily to members of the Army Reserve, though a few issues were made to the Royal Canadian Air Force for those who had prior service in the militia. This medal was replaced by the Canadian Forces' Decoration. It continued to be awarded in the United Kingdom until 2000 when it was replaced by the Voluntary Reserve Service Medal. The EM continues to be awarded in a number of other Commonwealth countries. It can be found with a wide variety of bars, from larger countries such as Australia to small territories such as Fiji, Hong Kong, Gibraltar, Guernsey, et cetera.
Criteria: Awarded to warrant officers and non-commissioned officers and men who completed twelve years of efficient service on the active list of the militia (the Army Reserve). War service counted as double toward the EM. A bar was awarded for each six years of additional service.
Efficiency Medal Elizabeth II 2nd type "Canadian" obverse.
Efficiency Medal reverse.
Insignia: A silver oval medal 39 mm tall and 32 mm wide; the obverse bears one of four different effigies:
* • George V issue: Crowned in coronation robes circumscribed by the text GEORGIVS — V — D — G — BRITT — OMN REX — ET — INDIAE — IMP —.
* • George VI 1st type: Crowned in coronation robes circumscribed by the text GEORGIVS — VI — D — G — BR OMN — REX — ET — INDAE — IMP —.
* • George VI 2nd type: Crowned in coronation robes circumscribed by the text GEORGIVS VI DEI GRA BRITT: OMN: REX FID:DEF:.
* • Elizabeth II 1st type: Wearing a Tudor Crown circumscribed by the text ELIZABETH II DEI GRATIA REGINA FID. DEF.
* • Elizabeth II 2nd type: Wearing a St. Edward's Crown circumscribed by the text ELIZABETH II DEI GRATIA REGINA FID. DEF. This obverse was only used in Canada.
The reverse of all issues bears the text FOR EFFICIENT SERVICE.
Suspender: A scroll bar linked to the medal by two laurel leaves. The scroll bears the name of the dominion or colony in which the medal was awarded. CANADA was the text on those medals awarded to Canadians.
Ribbon: A 32 mm wide green ribbon edged with 2 mm of yellow on each side.
Bars: A bar embossed with an Imperial Crown in the centre. Two holes are drilled in the side of each bar so that it can be sewn onto the ribbon.
Naming: Impressed or engraved capitals, abbreviated rank, initials, surname, and abbreviated unit.
Postnominals: None.
Other: This medal was normally issued in a white cardboard box bearing a crown on the lid.
Number Awarded to Canadians: 12,675 (1,600 George V, 10,600 George VI, 475 Elizabeth II), along with 3,200 1st bars, 845 2nd bars, 140 3rd bars, 18 4th bars, and 3 5th bars.
The Royal Canadian Navy Long Service and Good Conduct Medal
Royal Canadian Navy Long Service and Good Conduct Medal George V obverse.
Royal Canadian Navy Long Service and Good Conduct Medal George VI 2nd type obverse.
Origins: The Royal Canadian Navy Long Service and Good Conduct Medal was based directly upon the Royal Navy Long Service and Good Conduct Medal first instituted by King William IV on August 24, 1831. The original Royal Navy issue was awarded for twenty-one years of exemplary conduct. This term was reduced to ten years in 1874, but was later increased to fifteen years. Bars for fifteen years of additional service were instituted by King George V. After the creation of the Royal Navy Canadian Volunteer Reserve, which became the Royal Canadian Navy, this medal commenced being awarded in Canada.
Royal Canadian Navy Long Service and Good Conduct Medal Elizabeth II 1st type obverse.
Royal Canadian Navy Long Service and Good Conduct Medal reverse.
Criteria: Fifteen years of long service and good conduct in the Royal Canadian Navy. Service in Commonwealth navies could be counted toward earning this medal.
Insignia: A circular silver medal 36 mm in diameter. Seven different obverse types were awarded to Canadians:
* • George V 1st type: In admiral's uniform circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND IMP:.
* • George V 2nd type: Bareheaded circumscribed by the text GEORGIVS V BRITT: OMN: REX: ET IND: IMP:.
* • George VI 1st type (1937–48): Uncrowned circumscribed by the text GEORGIVS VI D:G:BR:OMN:REX F:D: IND: IMP:.
* • George VI 2nd type (1948–53): Uncrowned circumscribed by the text GEORGIVS VI D:G: BRITT: OMN: REX FID: DEF:.
* • Elizabeth II 1st type (1953–54): Uncrowned effigy circumscribed by the text ELIZABETH II DEI GRA: BRITT: OMN: REGINA: F:D:+.
* • Elizabeth II 2nd type (post-1954): Uncrowned effigy circumscribed by the text ELIZABETH II DEI GRATIA REGINA . F:D +.
The reverse bears Lord Nelson's HMS Victory circumscribed by the text FOR LONG SERVICE AND GOOD CONDUCT.
Suspender: A claw-footed suspender with a straight bar. Both swivelling and non-swivelling examples exist.
Ribbon: A blue ribbon 32 mm wide edged with 3 mm stripes of white.
Bars: A narrow bar decorated with laurel leaves.
Postnominals: None.
Naming: Engraved or impressed capitals, abbreviated rank, initials, surname, and RCN.
Number Awarded to Canadians: 864 (63 George V 1st type, 52 George V 2nd type, 315 George VI 1st type, 190 George VI 2nd type, 124 Elizabeth II 1st type, 120 Elizabeth II 2nd type).
The Royal Canadian Navy Volunteer Reserve Decoration/Royal Canadian Navy (Reserve) Decoration
Royal Canadian Navy Volunteer Reserve/Royal Canadian Navy (Reserve) Decoration obverse.
Origins: Following the pattern set by the Efficiency Decoration and its predecessor, the Volunteer Decoration, the Admiralty decided to institute a long service decoration for officers who served in the Royal Canadian Navy Volunteer Reserve (RCNVR). After the merging of the RCNVR and the Royal Canadian Navy Reserve (RCNR) into the Royal Canadian Navy (Reserve) or RCN(R), this decoration was retained for those serving in the RCN(R).
Criteria: Twenty years of service in the RCNVR. Service in a Commonwealth army, navy, or air force could be counted toward this award. This decoration was only awarded to those who enlisted before September 1, 1939. The first issues were made in 1938 and the last on September 1, 1959.
Insignia: A silver oval medal 54 mm tall and 33 mm wide. The outer edge consists of a loop of cable tied at the bottom in a reef knot and surmounted by a gilt Tudor Crown, the centre bearing the Royal Cypher of King George VI (GRI) in gilt openwork. The reverse is plain.
Suspender: A small loop is soldered onto the back of the crown. Through this passes a wider ring and the ribbon.
Ribbon: A 38 mm wide ribbon of dark blue defaced by a 6 mm wide green strip in the centre bordered by 3 mm stripe of red.
Bars: No bars were awarded to this decoration.
Naming: Engraved capitals, abbreviated rank, initials, surname, and RCNVR.
Postnominals: VRD.
Number: 52 (36 to the RCNVR and 16 to the RCN[R]).
The Royal Canadian Navy Reserve Decoration
Royal Canadian Navy Reserve Decoration obverse.
Origins: Following the pattern set by the Efficiency Decoration and its predecessor, the Volunteer Decoration, the Admiralty decided to institute a long service decoration for officers who served in the RCNR. In 1946 the RCNVR and the RCNR were merged into the RCN(R) and the awarding of the Royal Canadian Navy Reserve Decoration ceased, it being superseded by the Volunteer Reserve Decoration.
Criteria: Twenty years of service in the RCNR. Service in a Commonwealth army, navy, or air force could be counted toward this award. This decoration was only awarded to those who enlisted before September 1, 1939. The first issues were made in 1938 and the last in 1946.
Insignia: Identical to the RCNVR Decoration.
Suspender: Identical to the RCNVR Decoration.
Ribbon: A green ribbon 38 mm wide edged on each side by 2 mm of white. From 1937 to 1941 the ribbon was solid green and 38 mm wide. This did not show up well on the naval uniform and was easily confused with the Volunteer Decoration, so white stripes were added.
Bars: None.
Postnominals: RD.
Naming: Engraved capitals, abbreviated rank, initials, surname, and RCNR.
Number: 37.
The Royal Canadian Naval Volunteer Reserve Long Service and Good Conduct Medal/Royal Canadian Navy (Reserve) Long Service and Good Conduct Medal
Origins: This medal was based on the Royal Navy Long Service and Good Conduct Medal.
Criteria: Awarded to chief petty officers, petty officers, and men for twelve years of service in the RCNVR with the maximum number of good conduct badges from 1938 until the end of 1945 and service in the RCN(R) after January 1, 1946. This medal ceased to be awarded in September 1951. In order to be eligible for it a person had to have enlisted in the RCNVR before September 1, 1939. Bars were awarded for twelve years of additional service.
RCNVR Long Service and Good Conduct Medal/RCN(R) Long Service and Good Conduct Medal George VI 2nd type obverse.
Insignia: A circular silver medal 36 mm in diameter. The obverse depicts an uncrowned effigy of King George VI and is circumscribed by the text GEORGIVS VI D:G: BR: OMN: REX F:D: IND: IMP. The reverse depicts the starboard side view of the HMS Dreadnought with the Latin text DIUTURNE FIDELIS ("For long and faithful service").
Suspender: A claw-footed suspender with a straight bar.
Ribbon: A 32 mm wide ribbon with a dark green central stripe bordered by 6 mm stripes of navy blue and 2 mm of red.
RCNVR Long Service and Good Conduct Medal/RCN(R) Long Service and Good Conduct Medal reverse.
Bars: A narrow bar decorated with laurel leaves.
Naming: Engraved or impressed capitals, abbreviated rank, initials, surname, and RCNVR.
Postnominals: None.
Number: 601 (582 to the RCNVR, 19 to the RCN[R]) along with 19 bars, all to the RCNVR.
The Royal Canadian Naval Reserve Long Service and Good Conduct Medal
RCNR Long Service and Good Conduct Medal ribbon.
Origins: Based on the Royal Navy Long Service and Good Conduct Medal.
Criteria: Awarded to chief petty officers, petty officers, and men for fifteen years of service. This medal was discontinued in 1946 when the RCNR was merged with the RCNVR to form the RCN(R). Bars could be awarded for an additional fifteen years of service, but none were awarded to Canadians.
Insignia: Identical to the RCNVR Long Service and Good Conduct Medal/RCN(R) Long Service and Good Conduct Medal.
Suspender: A claw-footed suspender with a straight bar.
Ribbon: Originally the ribbon was solid green and 32 mm wide. Since this did not show up well on naval uniforms and conflicted with the old Colonial Auxiliary Long Service Medal, it was changed in 1941 to a dark green ribbon 32 mm wide and edged with 2 mm of white and a central 2 mm stripe of white.
Bars: A narrow bar decorated with laurel leaves.
Naming: Engraved or impressed capitals, abbreviated rank, initials, surname, and RCNR.
Postnominals: None.
Number: 54.
The Royal Canadian Air Force Long Service and Good Conduct Medal
RCAF Long Service and Good Conduct Medal George VI 2nd type obverse.
Origins: This medal was based on the Royal Air Force Long Service and Good Conduct Medal created by King George V on July 1, 1919, shortly after the establishment of the Royal Air Force. It was used in Canada from 1944 and was replaced by the Canadian Forces' Decoration in 1951, though members of the RCAF continued to be awarded it until 1957 if they had enlisted in the RCAF prior to September 1, 1939. Members of the RCAF had previously been awarded the Permanent Overseas Forces Long Service and Good Conduct Medal and then the Canadian Army Long Service and Good Conduct Medal.
Criteria: Awarded to permanent members of the RCAF (warrant officers, non-commissioned officers, and airmen) for eighteen years of service. Awarded for service between August 1, 1944, and September 1, 1957, providing the member was enlisted in the RCAF prior to September 1, 1939. In the United Kingdom, officers with twelve years in the ranks became eligible for the medal in 1947, though this policy had been followed by the RCAF since 1935 in Canada.
RCAF Long Service and Good Conduct Medal reverse.
Insignia: A circular silver medal 36 mm in diameter. The medal was issued with five different obverse types, though neither the King George V type nor the Elizabeth II 2nd type was awarded to members of the RCAF:
* • King George VI 1st type: Uncrowned effigy of George VI circumscribed with the text GEORGIVS VI D:G: BR: OMN: REX ET IND: IMP.
* • King George VI 2nd type: Uncrowned effigy of George VI circumscribed with the text GEORGIVS VI D:G: BRITT: OMN: REX FID: DEF:.
RCAF Long Service and Good Conduct Medal Elizabeth II obverse.
* • Queen Elizabeth II: Uncrowned effigy of Elizabeth II circumscribed with the text ELIZABETH II DEI GRATIA REGINA F:D:+.
The reverse depicts an eagle in flight surmounted by a Tudor Crown and circumscribed with the text FOR LONG SERVICE AND GOOD CONDUCT.
Suspender: A non-swivelling claw-footed suspender with an ornate scroll bar.
Ribbon: A 32 mm wide ribbon edged with 2 mm of white with the centre divided into half dark blue and half maroon.
Bars: A bar bearing a crown in the centre. None were awarded to Canadians.
Naming: Engraved capitals, abbreviated rank, initials, surname, and occasionally RCAF.
Postnominals: None.
Other: The medal continues to be awarded to members of the Royal Air Force. The period of service was reduced to fifteen years in 1977.
Number: 487 (primarily George VI 1st type).
The Air Efficiency Award
Air Efficiency Award George VI 1st type obverse.
Air Efficiency Award reverse.
Origins: Established on August 17, 1942, by Royal Warrant, the Air Efficiency Award is perhaps one of the most attractive long service awards in the Commonwealth. The medal was created to recognize "officers and airmen of any Auxiliary or Volunteer Air Force raised in Our United Kingdom... [India and the Colonies] and Our Dominions." As a dominion, Canada participated in the program, and members of the Royal Canadian Air Force Auxiliary qualified for the award. This was a novel concept because both officers and men shared in the same award, which was why it was referred to as an "award" rather than as either a "decoration" or a "medal" and why only officers could make use of the postnominals.
Criteria: Ten years of service of which not less than five years of actual service must have been spent in the RCAF Auxiliary. War service did not count doubly for the Air Efficiency Award as it did for the Efficiency Decoration and Efficiency Medal.
Insignia: An oval-shaped silver medal 38 mm in height and 32 mm in width. The obverse depicts an uncrowned effigy of King George VI and his titles GEORGIVS VI D:G BR: OMN: REX F:D:IND:IMP:. The reverse is plain and bears only the text AIR EFFICIENCY AWARD on three lines. None of the later version of this medal was awarded to Canadians.
Suspender: An eagle with wings extended, with a straight bar attached to the wing tips through which the ribbon passes.
Ribbon: A 34 mm wide ribbon green in colour divided in the centre by two equal stripes of pale blue 2 mm in width spaced 2 mm apart.
Bars: A bar defaced in the centre by an eagle with wings extended. The bar has two holes in each corner so that it can be sewn onto the ribbon. No bars were awarded to Canadians.
Postnominals: AE (for officers only).
Naming: Impressed or engraved capitals, abbreviated rank, initials, and surname.
Other: Canadian officials originally intended for this award to have a bar with CANADA on it, but that never happened.
Number: 94.
The Imperial Service Medal
Imperial Service Medal George V star issue obverse.
Imperial Service Medal Edward VII star issue obverse.
Origins: The Imperial Service Medal (ISM) is most often associated with the Imperial Service Order (ISO), both awards having been established by King Edward VII on August 8, 1902, and by means of the same Royal Warrant. Initially, women were not permitted to receive the medal, but in 1908 the regulations were changed to allow for this. While the ISO was primarily reserved for senior civil servants, the ISM was for more junior clerks and labourers who worked in various government departments. The Imperial Service Medal was the only long service medal awarded to civil servants and was fairly broadly bestowed.
Awards of the Imperial Service Medal to Canadians ceased in late 1952. The last issues were all George VI 2nd type. No Queen Elizabeth II issues were awarded to Canadians. In Canada the program was administered through the Department of the Secretary of State. Medals were usually posted to the recipients, though there are a number of instances for which investitures were held by local officials. It was not possible to be awarded both the Imperial Service Medal and the Imperial Service Order.
Imperial Service Medal George V uncrowned issue obverse.
The medal continues to be awarded in a few of the Queen's realms. The last Australian awards came in 1989 and the last British awards were made in 1995. The total number of awards of the medal from 1902 to the present is in excess of a quarter million.
Criteria: The medal was awarded to "those who are not eligible for appointment as Companions of the Order." Workers qualified "on retirement from Our said Services after not less than twenty-five years of meritorious service in the United Kingdom, or not less than sixteen years of such service in a colony or protectorate." Canada was not considered an "unhealthy" place, thus, the medal was awarded for twenty-five years of meritorious service in Canada. In Canada the ISM was only awarded to "outside" workers such as postmen, postmistresses, railway switch operators, canal lock men, and other related occupations.
There were two separate insignia styles: the "star" shape and the circular medal shape. The star-shaped insignia was replaced by the circular silver medal on June 25, 1920.[2] The star shape is identical in appearance to the Imperial Service Order. However, it was made of bronze and silver, not silver and gold as the order is.
Insignia: For the star-shape issues, a six-pointed bronze star 56 mm wide, in the centre of which is a silver disc bearing the Royal Cypher (EVIIR, GVR, or GRI) circumscribed with FOR FAITHFUL SERVICE in blue enamel surmounted by an Imperial Crown. The circular silver medal was 32 mm in diameter. Four separate effigies were used on this medal:
Imperial Service Medal reverse.
Imperial Service Medal George VI 1st type obverse.
* • George V 1st type: Uncrowned head circumscribed by GEORGIVS V. D.G. BRITT OMN REX F.D. IND : IMP:.
* • George V 2nd type: Crowned head circumscribed by GEORGIVS V D.G. BRITT. OMN : REX. ET. INDIAE. IMP.
Imperial Service Medal certificate, 1922.
Imperial Service Medal certificate, 1932.
* • George VI 1st type: Uncrowned head circumscribed by GEORGIVS VI D: G: BR: OMN: REX ET INDIAE IMP:.
* • George VI 2nd type: Uncrowned head circumscribed by GEORGIVS VI DEI: GRA: BRITT OMN: REX FID DEF +.
Imperial Service Medal certificate, 1947.
The reverse of all of the circular medals displays an unclothed man seated and resting from his labours with the text FOR FAITHFUL SERVICE at the base.
Suspender: Star issues have a bronze ring attached to the top of the crown; circular issues have a claw-footed suspender with a ring 15 mm in diameter.
Ribbon: A 38 mm wide ribbon made up of three equal parts of watered red, blue, and red.
Bars: None.
Naming: Star issues were engraved on the reverse with the given and surname in capitals. Circular issues were issued with engraved or impressed capitals, full given names, and surname.
Postnominals: None.
Other: All medals were awarded with a certificate. There were several types of certificates. The first two types were printed on British Home Office stationery, with the date of award written in cursive or typed in the upper-right-hand corner, and the name of the recipient similarly entered at the bottom of the certificate. The second type of certificate was slightly more elaborate in that it included the seal of the Imperial Service Order embossed in the upper-left-hand corner.
Number: 7,095.[3]
# 8
IMPERIAL COMMEMORATIVE MEDALS
The tradition of presenting loyal subjects with commemorative medals can be traced back more than half a millennium to India's Pudukkottai state where khelats were presented on special occasions. These khelats were not high honours but emblems of recognition and also served as a means to ensure stability within the existing order.[1] While the bestowal of commemorative medals may have this effect today, it is certainly no longer the intention of such awards. They are, in effect, small tokens of gratitude given by the sovereign on special occasions to deserving people both civilian and military. In addition to the seven medals listed in this chapter, the British also awarded medals on the occasion of the Delhi Durbar in 1903 and 1911. No awards of these medals were made to people living in Canada at the time, though there are a few instances where officers serving in the Imperial Army or Navy received these medals and later moved to Canada.
The British also awarded special Mayors' and Provosts' Medals for Queen Victoria's 1897 Diamond Jubilee and King Edward VII's coronation in 1902. No Canadian mayors received these medals because they were reserved for officials in the United Kingdom.
All of the commemorative medals examined in this chapter were awarded to British subjects throughout the British Empire from Australia to Zanzibar, though awards to some dominions and colonies are quite rare. Aside from the United Kingdom, Canada — as the largest dominion — received the largest allotment of medals, followed closely by India and Australia.
Only a small number of Canadians received the early commemorative medals. Occasionally, Queen Victoria Diamond Jubilee, King Edward VII Coronation, and King George V Coronation Medals are found privately named. No postnominal designations are attached to any of these awards. Until the 1935 Silver Jubilee of King George V, a person had to actually attend or be involved in the official jubilee or coronation ceremonies in order to receive a medal. Others received it because of the high office they held.
With the 1935 Silver Jubilee, it was decided that for future royal events such as coronations or jubilees, medals would be awarded to a wider array of people during the event year rather than just those present for the event itself. With this, the distribution of commemorative medals became more democratic.
Rolls of the Canadian recipients exist for the 1887, 1897, and 1911 jubilee/coronation medals. A complete listing of Canadian recipients of the King George V Silver Jubilee Medal was published in the Canada Gazette on May 4, 1935. This was the only occasion, prior to the creation of the modern honours system, when a complete list of recipients was made public. Special souvenir booklets were also published, detailing the Canadian military participants who attended the ceremonies in Britain.
King Charles II Coronation Medal obverse.
King Charles II Coronation Medal reverse.
Aside from war medals awarded for service in the First and Second World Wars, it was coronation/jubilee medals that first introduced Canadians to the honours system. Not restricted to members of the armed forces alone, many average citizens received recognition, beginning with the King George V Silver Jubilee Medal and continuing to this day. These were also the first medals to be awarded in great number to women. The King George V Silver Jubilee Medal was bestowed upon a wide cross-section of Canadian society, including a significant number of women and aboriginal Canadians.
The Queen Victoria Golden Jubilee Medal (1887)
Queen Victoria Golden Jubilee Medal in silver obverse.
Queen Victoria Golden Jubilee Medal in silver reverse.
Origins: This was the first standardized issue of a wearable jubilee/commemorative medal in the Commonwealth. To commemorate the fiftieth anniversary of Queen Victoria's accession to the throne, the British government decided to issue a special medal to commemorate the occasion. The noted die engravers J.B. and A.B. Wyon were commissioned to design the medal. This was the first time that a jubilee medal was hung from a ribbon and intended to be worn with military medals. The medal was struck in gold, silver, and bronze.
Gold issues were presented to Queen Victoria and various members of the Royal Family as well as to certain favoured Indian princes. Silver issues were presented to officers and some government officials. The bronze issues were reserved for non-commissioned officers and junior government officials.
Criteria: Medals were primarily presented to those officials who attended the jubilee celebrations in London, but a significant number were also bestowed upon those in positions of authority.
Insignia: A circular medal 30 mm in diameter struck in bronze, silver, or gold, depicting a veiled Queen Victoria on the obverse circumscribed by the text VICTORIA D.G. REGINA ET IMPERATIX. F.D. The reverse is surrounded by a wreath of roses, thistles, shamrocks open at the top to include a Royal Crown and tied at the base by a ribbon. In the centre of the reverse is the text on eight lines: IN COMMEMORATION OF THE 50TH YEAR OF THE REIGN OF QUEEN VICTORIA 21 JUNE 1897.
Suspender: An eyelet screwed into the top of the medal through which passes a ring and the ribbon.
Ribbon: A 32 mm wide blue ribbon defaced with two 5 mm white stripes set 2 mm from the edges.
Bars: None to Canadians, though bars were awarded to those involved in the 1897 Diamond Jubilee. The bar bears the date 1897 surrounded by a rope and surmounted by a Royal Crown.
Other: The medals were struck by Wyon & Co.
Number: 361 gold, 1,235 silver, and 5,859 bronze medals were struck. Two of the silver medals were awarded to Canadians.
The Queen Victoria Diamond Jubilee Medal (1897)
Queen Victoria Diamond Jubilee Medal in silver reverse.
Queen Victoria Diamond Jubilee Medal in silver obverse.
Origins: In light of the success of the Golden Jubilee Medal, it was decided to issue a subsequent medal to commemorate the Diamond Jubilee of Queen Victoria's accession to the throne.
Criteria: As with the Golden Jubilee Medal, the Diamond Jubilee Medals were primarily presented to those officials who attended the jubilee celebrations in London, but a significant number were also bestowed upon those in authority. Medals were struck in gold, silver, and bronze. Members of the special Canadian contingent were awarded the medal in bronze and silver. These medals were personally presented by His Royal Highness The Prince of Wales (the future King Edward VII) at Buckingham Palace on July 3, 1897.
Insignia: A circular medal 30 mm in diameter struck in gold, silver, or bronze. The obverse bears a crowned and veiled bust of Queen Victoria and is circumscribed by the text VICTORIA D.G. REGINA ET IMPERATIX F.D. The reverse is surrounded by a wreath of roses, thistles, and shamrocks open at the top to include a Royal Crown and tied at the base by a ribbon. In the centre of the reverse is the text on eight lines: IN COMMEMORATION OF THE 60TH YEAR OF THE REIGN OF QUEEN VICTORIA 20 JUNE 1897.
Suspender: An eyelet screwed into the top of the medal through which passes a ring and the ribbon.
Ribbon: A 32 mm wide blue ribbon defaced with two 5 mm white stripes set 2 mm from the edges. This is identical to the ribbon issued with the 1887 Golden Jubilee Medal.
Bars: None.
Other: The medal was designed and struck by A.B. Wyon. It is essentially a slightly modified version of the 1887 Golden Jubilee Medal. Recipients of the 1887 Golden Jubilee Medal were not permitted to receive this award and instead were presented with a bar bearing the date 1897.
Number: 73 gold, 3,040 silver, and 890 bronze were issued; 37 silver and 165 in bronze to Canadians.
The King Edward VII Coronation Medal (1902)
King Edward VII Coronation Medal in bronze obverse.
King Edward VII Coronation Medal in bronze reverse.
Origins: The first standardized coronation medal intended for wear, this medal followed in the tradition of the 1887 and 1897 jubilee medals that were awarded to commemorate the various Jubilees of Queen Victoria's accession to the throne. The design of this medal was in part taken from the Royal Victorian Medal.
Criteria: Primarily presented to those officials who attended the jubilee celebrations for Edward VII in London, but a significant number were also bestowed upon those in authority. Medals were struck in gold, silver, and bronze. Members of the special Canadian contingent were also awarded the medal in bronze and silver — silver to officers and bronze to non-commissioned officers.
Insignia: A circular medal 30 mm in diameter struck in gold, silver, or bronze. The obverse contains the conjoined busts of King Edward VII and Queen Alexandra surrounded by a thin border of laurel leaves and surmounted by a Tudor Crown. The reverse contains the Royal Cypher EVIIR and the date 26 JUNE 1902. The coronation was actually postponed because King Edward VII was ill; the medals, having already been struck, were issued with the original and not the actual date of the coronation.
Suspender: A small ring was attached to the crown on the top of the medal.
Ribbon: A dark blue ribbon 32 mm wide edged with 2 mm of white and bearing in the centre a 7 mm wide red stripe.
Bars: None.
Other: Medals struck by Elkington & Company.
Number: 3,494 silver and 6,054 bronze were issued; 25 silver and 608 bronze to Canadians.
The King George V Coronation Medal (1911)
King George V Coronation Medal obverse.
Origins: After the tradition established by King Edward VII in 1902, a special coronation medal was established to commemorate the coronation of King George V. As with the coronation medal of Edward VII, separate issues were struck for members of the various British police services, St. John Ambulance, and a variety of other quasi-government organizations. Little more than a year later a special medal was struck for the Delhi Durbar at which King George V was crowned emperor of India. This medal uses the same ribbon as the King George V Coronation Medal, and no recipient was awarded both the Delhi Durbar Medal and the George V Coronation Medal. This was the first coronation/jubilee medal issue in which the medal was struck in silver alone. Thus, men, non-commissioned officers, officers, and princes all received the same medal.
King George V Coronation Medal reverse.
Criteria: Participation in the coronation ceremonies/celebrations or some other activity that was deemed worthy of recognition by the British government.
Insignia: A circular silver medal 32 mm in diameter. The obverse depicts the conjoined busts of King George V and Queen Mary surrounded by a floral wreath. The reverse has a beaded border and is plain bearing only the Royal Cypher of King George (GVR) and the date 22 JUNE 1911.
Suspender: An eyelet screwed into the top of the medal through which passes a ring and the ribbon.
Ribbon: A blue ribbon 32 mm wide defaced in the centre by two 2 mm red stripes set 1 mm apart.
Bars: None.
Other: Medals struck at the Royal Mint.
Number: 15,901 medals were struck; 1,007 to Canadians.
The King George V Silver Jubilee Medal (1935)
King George V Silver Jubilee Medal obverse.
Origins: Following the tradition established in 1887 with the golden jubilee of Queen Victoria's accession to the throne, it was decided to strike a medal to commemorate the twenty-fifth anniversary of King George V's accession to the throne. This would become the broadest issue of a commemorative medal until the Canadian Centennial Medal was struck in 1967, and it was the first time that Canadian MPs and senators were given an allotment of medals to nominate constituents. Indeed, it was with the 1935 Silver Jubilee Medal that the tradition of allowing local representatives to select worthy recipients was initiated.
Criteria: Participation in the official silver jubilee celebrations or being selected by a senior government official, senator, or MP as having performed valuable services to the community or Canada as a whole. On occasion they were used for partisan purposes to reward loyal party workers.
Insignia: A circular silver medal 32 mm in diameter. The obverse bears the conjoined busts of King George V and Queen Mary circumscribed by the text GEORGE V AND QUEEN MARY MAY IV MCMXXXV. The reverse contains the Royal Cypher GRI and the dates MAY 6 1910 — MAY 6 1935.
Suspender: A claw-footed suspender with a ring 12 mm in diameter.
Ribbon: A dark red ribbon 32 mm in width edged on each side by 1 mm of blue, 2 mm of white, and 1 mm of blue.
Bars: None.
King George V Silver Jubilee Medal reverse.
King George V Silver Jubilee Medal certificate.
Other: Recipients also received certificates with their names (typewritten or in calligraphy) on watermarked paper bearing the Royal Arms in red at the top.
Number: 85,324 for the entire British Empire; 7,500 to Canadians.
The King George VI Coronation Medal (1937)
King George VI Coronation Medal obverse.
King George VI Coronation Medal reverse.
Origins: The King George VI Coronation Medal was modelled on the 1911 King George V Coronation Medal and was awarded to commemorate the coronation of King George VI.
Criteria: Participation in the official coronation celebrations or being selected by a government official, senator, or MP as having performed valuable services to the community or Canada as a whole.
Insignia: A silver circular medal 32 mm in diameter. The obverse bears the conjoined busts of King George VI and Queen Elizabeth. The reverse is defaced by the Royal Cypher GRI in the centre and the text CROWNED 12 MAY 1937 below and circumscribed by the text GEORGE VI QVEEN ELIZABETH.
King George VI Coronation Medal certificate.
Suspender: Claw-footed suspender with a ring 12 mm in diameter.
Ribbon: A blue ribbon 32 mm wide edged on each side by 1 mm of white, 2 mm of red, and 1 mm of white.
Bars: None.
Other: Recipients received certificates bearing their names (typewritten or in calligraphy) on watermarked paper bearing the Royal Cypher in red at the top.
Number: 90,279 were struck; 10,359 to Canadians.
Queen Elizabeth II Coronation Medal (1953)
Queen Elizabeth II Coronation Medal obverse.
Queen Elizabeth II Coronation Medal reverse.
Origins: The Queen Elizabeth Coronation Medal was struck to commemorate the coronation of Queen Elizabeth II and was part of the long-established tradition of awarding a special commemorative medal to citizens and members of the armed forces on the occasion of a coronation or jubilee. This was the last Commonwealth-wide issue of a coronation medal.
Criteria: Participation in the coronation or being selected by a government official, senator, or MP as having performed valuable services to the community or Canada as a whole.
Insignia: A circular silver medal 32 mm in diameter. The obverse depicts a crowned bust of Queen Elizabeth II. The reverse contains the Royal Cypher EIIR in the centre and contains the text QUEEN ELIZABETH CROWNED 2ND JUNE 1953.
Queen Elizabeth II Coronation Medal certificate.
Suspender: A claw-footed suspender with a ring 12 mm in diameter.
Ribbon: A dark red ribbon 32 mm in width, with two 2 mm stripes of blue set 1 mm apart in the centre and edged by 1 mm of white.
Bars: None.
Other: Recipients received certificates bearing their names (typewritten or in calligraphy) on watermarked paper bearing the Royal Cypher in red at the top.
Number: 129,000 struck; 19,000 to Canadians.
Queen Elizabeth II investing Commander R.P. Welland, DSC, CD, RCN, with the Coronation Medal at Buckingham Palace.
# 9
SPECIAL IMPERIAL MEDALS
There were a variety of imperial awards bestowed upon residents of Canada that do not fit into the standard categories for medals. These awards include medals given for Arctic exploration and marksmanship. Rather coincidentally, these medals were ranked at the very end of the Order of Precedence, which is a bit odd in the case of the Arctic and Polar Medals since they were earned as the result of great hardship. None of these medals carry a postnominal designation, and the number awarded to Canadians was very small.
The Arctic Medal (1818–55)
Origins: European exploration of the Arctic began in the sixteenth century, most specifically in the quest for a speedy route to India. A number of great battles were fought on the waters of Hudson Bay, and the region has been subject to numerous scientific and military expeditions.
Criteria: The medal was awarded to all men and officers and employees of the Hudson's Bay Company who participated in the various expeditions to the Arctic regions of what is today northern Canada.
Insignia: An octagonal silver medal 34 mm in diameter. The obverse bears the effigy of Queen Victoria circumscribed by VICTORIA REGINA. The reverse carries a depiction of a three-masted sailing ship amid icebergs and an ice jam with sailors and an ice sledge in the foreground with the inscription FOR ARCTIC DISCOVERIES at the top and the dates 1818–1855 at the bottom.
Arctic Medal 1818–55 obverse.
Arctic Medal 1818–55 reverse.
Suspender: A claw-footed suspender attached to a five-pointed Northern Star atop which is fitted a swivel-ling ring.
Ribbon: White moiré 38 mm in width.
Bars: None.
Naming: Issued unnamed; however, examples are found in various private hand-engraved styles.
Other: The medal was designed by the noted nineteenth-century engraver L.C. Wyon and struck by the Royal Mint in London. Because of its delicate swivelling ball and star suspension assembly, the medals were prone to break. It is not unusual to see Arctic Medals with repaired suspensions or occasionally an improvised straight suspension.
Number: 1,486 total, with 1,106 to the Royal Navy and 380 to Hudson's Bay Company staff and scientific staff from Britain and other countries.
The Arctic Medal (1875–76)
Arctic Medal 1875–76 obverse.
Origins: While some thought was given to reissuing the Arctic Medal (1818–55), the inclusion of dates on the reverse precluded this. A less-complicated design and suspender was devised and engraved by G.G. Adams and L.C. Wyon.
Criteria: Awarded to the officers and men of HMS Alert and HMS Discovery who served in the Arctic expedition that transpired between July 17, 1875, and November 2, 1876. The eligibility for the medal was later broadened to include those who sailed on the private yacht Pandora, which journeyed in polar waters between June 25 and October 19, 1875.
Insignia: A silver circular medal 36 mm in diameter. The obverse bears a crowned effigy of Queen Victoria with the text VICTORIA REGINA 1876 and a beaded border, while the reverse depicts an icebound vessel. The obverse effigy is almost identical to that used for the Empress of India Medal of 1877.
Arctic Medal 1875–76 reverse.
Suspender: A swivelling claw suspender with a straight bar.
Ribbon: White moiré 38 mm in width.
Naming: All Arctic Medals (1875–76) were named in impressed capitals. A few examples impressed with SPECIMEN are also known.
Bars: None.
Other: The medal was struck by the Royal Mint in London.
Number: 170; 57 to Discovery, 63 to Alert, and 50 to Pandora.
The Polar Medal
Polar Medal in silver Edward VII obverse.
Polar Medal in silver reverse.
Origins: The Polar Medal is the final in the series of medals awarded for polar exploration. This medal continues to be awarded for both Arctic and Antarctic explor-ation, though most are awarded for scientific work in the Antarctic. While it is still awarded to the British, it is no longer awarded to Canadians. This medal was replaced by the Canadian Polar Medal in 2015.
Criteria: Awarded to those participating on officially sanctioned expeditions to either the North or South Poles for the purposes of exploration or scientific research.
Insignia: Silver or bronze, octagonal in shape, and 33 mm in diameter. The bronze medal was discontinued in 1939. There are six obverse designs:
* • Edward VII: Uncrowned in an admiral's uniform circumscribed by the text EDWARDVUS VII REX IMPERATOR.
* • George V 1st type: Uncrowned George V in admiral's uniform circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:.
* • George V 2nd type: Crowned George V in coronation robes circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:.
* • George V 3rd type: Uncrowned coinage profile circumscribed by the text GEORGIVS V BRITT: OMN: REX ET IND: IMP:.
* • George VI: Depicting an uncrowned George VI circumscribed by the text GEORGIVS VI D:G:BR:OMN: REX: F:D: IND: IMP.
* • Elizabeth II: Uncrowned head circumscribed by the text ELIZABETH II DEI:GRA: BRITT: OMN: REGINA: F:D:+.
The reverse is common to all issues and depicts Royal Research Ship Discovery in the background with an ice-sledge party of eight men in the foreground.
Suspender: A claw-footed suspender with an ornate scroll bar.
Ribbon: White moiré 38 mm in width.
Bars: A silver bar with the name and date of the particular expedition is attached to the suspender of each medal. The bars, which are unique to each expedition, were orignally struck, however since 1941 they have been hand-engraved.
Naming: A variety of styles; most are hand-engraved in capitals, while some have been found to have impressed capitals.
Other: The medal was designed by E.G. Gillick and continues to be struck by the Royal Mint. This medal retains a design very similar to that of the Arctic Medal (1818–55).
Polar Medal in silver Elizabeth II obverse.
Number: 16 to Canadians for service to Canada, all in connection with Royal Canadian Mounted Police patrols of the Arctic during the Second World War.
* • ARCTIC 1940–42 and ARCTIC 1944: 3.
* • ARCTIC 1940–42: 5.
* • ARCTIC 1944: 8
(5 to Canadians working in connection with British expeditions to the Antarctic).
* • ANTARCTIC 1907–1909 in bronze: 1.
* • ANTARCTIC 1910–1913: 1.
* • ANTARCTIC 1944–45: 1.
* • ANTARCTIC 1950–52: 1.
* • ANTARCTIC 1985–86: 1.
The King's/Queen's Medal for Champion Shot of the Canadian Army
Origins: This medal has gone through a number of iterations, originally carrying the name the Army Best Shot Medal. It was instituted by Queen Victoria on April 30, 1869, and was awarded to members of the British Army for winning the annual army marksmanship contest held at Bisley in the United Kingdom. The medal was intended to encourage high proficiency in marksmanship and was originally struck in bronze but was changed to silver in 1872. The award became dormant in 1882.
In 1922 the regulations for the medal were revived and altered, opening it up to members of the various dominion armies for the first time. It was at this time that date bars were instituted, but it was not until 1923 that the medal was awarded to members of the Canadian Army when a formal Canadian competition was inaugurated.
The medal was eventually awarded to the military forces of Australia, Canada, Ceylon, India, Rhodesia, the British South Africa Police, the Union of South Africa, Pakistan, Jamaica, and Ghana. Canada was the last of these countries to continue awarding the medal.
With the unification of the Canadian Army into the Canadian Armed Forces (CAF) in 1968, the Queen's Medal for Champion Shot of the Canadian Army was discontinued on paper and replaced by what became unofficially known as the Queen's Medal for Champion Shot of the Canadian Forces, which was exactly the same medal.
King's Medal for Champion Shot of the Canadian Army George V 2nd type obverse.
King's Medal for Champion Shot of the Canadian Army reverse.
The zeal with which unification was imposed left nothing unchanged, even if it was only on paper. This was the last imperial medal to be awarded to Canadians in Right of Canada. In 1991 it was replaced by the Queen's Medal for Champion Shot Canada, which bears the same reverse allegory and an obverse that is unique to Canada.
Queen's Medal for Champion Shot of the Canadian Army obverse.
In addition to this medal there was also the Naval Good Shooting Medal, which was established by King Edward VII in 1902. This medal was only awarded between 1903 and 1914, with 974 being awarded during that short period. The medal was awarded for proficient gunnery performances in the annual Fleet Competitions. It was never awarded to members of the Royal Navy Canadian Voluntary Reserve because the force had barely come into existence when the medal was discontinued. Indeed, it remains an anomaly that the Royal Navy has yet to reinstitute a champion shot medal of any kind. In 1958, New Zealand established the Queen's Medal for Champion Shot of the New Zealand Naval Forces.
When inaugurated in 1923, one medal per year was awarded to the member of the regular Canadian Army who won the Best Shot Competition for Rifle Shooting at Bisley in the United Kingdom or at the Connaught Ranges near Ottawa. From 1953 to 1963, members of Canada's national police force, the Royal Canadian Mounted Police, participated in the competition, though none won the medal during this period.
In 1964 it was decided that two medals per year would be awarded, one for the regular army and the other for the reserve army and the RCMP. Upon unification of the Canadian Army into the CAF in 1968, two medals continued to be awarded annually, though the medal was now open to the entire CAF, with one medal going to regular members and the other going to reservists and members of the RCMP.
Criteria: Highest aggregate score achieved in the Best Shot Competition.
Insignia: A silver circular medal 36 mm in diameter. The reverse depicts the figure of Fame rising from her throne, facing left, with a horn in her left hand. With her right hand she is crowning, with a laurel wreath, a warrior facing right. The warrior's weight is on his right foot, since his left foot is raised and resting on the dais to support — on his left knee — a target having three arrows in the centre. In his right hand he holds a bow and a quiver full of arrows. The reverse was designed by Sir E.J. Poynter and engraved by the noted die engraver A.B. Wyon. There have been five different obverses for this medal, however it is uncertain whether the George VI 2nd type was bestowed upon a Canadian:
* • George V 1st type: Uncrowned wearing a field marshal's uniform and circumscribed by GEORGIVS V BRITT:OMN:REX ET INDIAE:IMP:.
* • George V 2nd type: Crowned wearing coronation robes and circumscribed by GEORGIVS V BRITT:OMM:REX ET INDIAE:IMP:.
* • George VI 1st type: Crowned and circumscribed by the text GEORGIVS VI D:G: BR: OMN: REX F:D: IND: IMP.
* • George VI 2nd type: Crowned and circumscribed by the text GEORGIVS VI DEI GRA BRITT: OMN: REX FID: DEF:.
* • Elizabeth II: Crowned effigy circumscribed by the text ELIZABETH II DEI GRATIA REGINA F.D:.
Suspender: Three types of suspension exist: a swivelling claw-footed suspender with straight bar, a non-swivelling claw-footed suspender with straight bar riveted to the medal, and a non-swivelling claw-footed suspender pinned to the top of the medal with a straight bar. It is believed that the suspension was last changed in 1965.
Ribbon: A maroon ribbon 32 mm wide edged with 3 mm of black, 3 mm of white, and 3 mm of black on each side.
Clasps: A silver clasp bearing the year of award in relief was awarded with each medal. The clasp is riveted to the suspender. This medal is always issued with a clasp of which it is possible to earn more than one.
Naming: Impressed or engraved capitals, abbreviated rank, initials, surname, and abbreviated unit.
Other: The medal was not awarded from 1940 to 1946 on account of the Second World War. Until 1952 medals were presented in a white cardboard box bearing a Tudor Crown on the lid. After 1952 the medals were presented in a Farrington case, the same case used for the Canadian Forces' Decoration.
Number: 69.
* • King George V: 9 medals with bar, 6 additional bars.
* • King George VI: 4 medals with bar, 4 additional bars.
* • Queen Elizabeth II: 56 medals with bar, 20 additional bars.
In an extraordinary feat, Major Desmond Thomas Burke managed to win the medal seven times (medal with six bars): 1925, 1927, 1929, 1930, 1931, 1939, and 1947. One woman, Private Shannon Wills, has been awarded the medal. She won the competition in 1988.
The Queen's Medal for Champion Shot of the RCAF
Queen's Medal for Champion Shot of the RCAF obverse.
Origins: This medal was modelled on the Army Best Shot Medal, which was instituted in 1869 and remains on issue for members of the British Army. The Queen's Medal for Champion Shot of the Royal Air Force was created on June 12, 1953, and the following year the Royal Canadian Air Force held its first shooting competition for the medal. Prior to this, the RCAF did not have an annual shooting competition. With the unification of the Royal Canadian Air Force into the Canadian Armed Forces in 1968, this medal ceased to be issued and all members of the Canadian Armed Forces became eligible for the Queen's Medal for Champion Shot of the Canadian Army (which was unofficially retitled Queen's Medal for Champion Shot of the Canadian Armed Forces in 1968). It was therefore only issued between 1954 and 1967.
Criteria: The medal was awarded following an annual competition held by the Dominion of Canada Rifle Association. Utilizing the same small arms championship conditions used for the Champion Shot of the Army Medal, the competition was held using the FN 7.62 mm service rifle. It was open to both regular and reserve members of the RCAF who achieved the best aggregate score in the competition.
Insignia: A silver circular medal 36 mm in diameter. The obverse bears a crowned effigy of Queen Elizabeth II circumscribed by the text ELIZABETH II D:G: BR: OMN: REGINA F:D:. The reverse shows an allegorical figure of Hermes mounted atop a hawk in flight holding in one hand a caduceus and the other erect preparing to throw a javelin. The reverse is circumscribed by the text THE QUEEN'S MEDAL FOR CHAMPION SHOTS OF THE AIR FORCES. When the medal was created in 1953, it was realized that including the name of the Royal Air Force would not be appropriate, since the medal was also to be awarded to members of the Royal Australian Air Force, Royal Canadian Air Force, Royal New Zealand Air Force, and South African Air Force (though none were awarded to the SAAF), hence the more generic inscription. The medal was designed by B.R. Sindall of the Royal Mint and continues to be struck there.
Suspender: Two types of suspension exist: a non-swivelling claw-footed suspender with straight bar is riveted to the medal or a swivelling claw-footed suspender is pinned to the top of the medal (no rivet through the face of the medal) with a straight bar. It is believed that the suspension was changed in 1965.
Queen's Medal for Champion Shot of the RCAF reverse.
Ribbon: A maroon ribbon 32 mm wide edged with 3 mm of dark blue, 3 mm of light blue, and 3 mm of dark blue on each side.
Clasps: A silver clasp bearing the year of award in relief was awarded with each medal. The clasp is riveted to the suspender. This medal is always issued with a clasp of which it is possible to earn more than one.
Naming: Impressed or engraved capitals, abbreviated rank, initials, surname, and RCAF.
Other: Medals were named with impressed capitals or hand-engraved block capitals. Medals were usually presented with a rectangular black leatherette case bearing the Canadian coat of arms on the lid. The identical medal continues to be awarded to members of the Royal Air Force.
Number: 12 (2 medals with two date clasps).
# 10
THE CONFEDERATION MEDAL: CANADA'S FIRST HONOUR
The Confederation Medal is unusual among Canadian awards/honours because of its massive size and the manner in which it was bestowed.[1] The idea for the medal seems to have come from none other than Sir John A. Macdonald, Canada's first prime minister. A quote for the sinking of the dies and striking the medal in gold, silver, and bronze was given to Macdonald on April 16, 1867, several months before Confederation was due to take place. Macdonald did not act immediately on this, and it was not until June 3, 1868, that the medal was authorized by Order-in-Council 1868-598 and Queen Victoria's permission was sought.
> On a Memorandum dated 3rd June 1868, from the Hon. The Minister of Justice and Attorney General, submitting that the Delegates to England on the subject of Confederation in 1867, thought it well that a medal should be struck commemorating the Confederation of the Provinces — he, therefore at their desire made arrangements for the preparation of the medal with Mess'rs J.S. and A.B. Wyon Chief Engravers of Her Majesty's Seals.
>
> That Her Majesty was graciously pleased to approve of the design and the medal is now being prepared; he therefore, recommends that a remittance be made to the Engravers of two thousand dollars, the sum voted for that object in the estimates for 1869, and that they be instructed to prepare one Gold, fifty Silver and five hundred Bronze Medals, without delay.
>
> The Committee advise that the recommendation of the Hon. Minister of Justice be approved and acted on.
>
> John A. Macdonald[2]
Drawing of Confederation Medal.
The medal was designed and struck by J.S. and A.B. Wyon, master engravers who were responsible for most of the Royal Seals of the period, including the first Great Seal of the new Dominion of Canada. The medals struck in silver were viewed as awards of merit for the Fathers of Confederation, while those in bronze were viewed more as commemorative medals marking the recipient's role in the first legislatures of the new country.
The first allotment of Confederation Medals was distributed between 1868 and 1869, with the one gold issue going to Queen Victoria, sixteen silver medals going to the sixteen delegates to the London Conference, four additional silver medals to members of the federal Cabinet who were participants in the London Conference, one for the Prince of Wales (the future Edward VII), one for Viscount Monck, the governor general, one for the colonial secretary, three to the provincial premiers, and one to the Library of Parliament. Bronze medals were given to each of the 181 Members of the House of Commons and the seventy-two members of the Senate, delegates who attended the Charlottetown and Quebec Conferences and who did not fit into these other categories, members of each of the provincial legislative councils and legislative assemblies, and single examples were deposited with seventy-five institutions, mainly educational in nature.
A second distribution of Confederation Medals took place in 1872 when eleven more silver Confederation Medals and four bronze were distributed. By 1885 the entire order of fifty silver medals had been exhausted.
Bronze Confederation Medal obverse.
Bronze Confederation Medal reverse.
Additional awards of the medal in bronze were made from 1885 through to 1894, usually to ministers of the Crown. The last Confederation Medal to be presented to an individual was given on the advice of the Honourable Fernand Rinfret, secretary of state, on July 1, 1927, to Prime Minister William Lyon Mackenzie King. In a typically ingratiating letter to Mackenzie King, Rinfret noted: "Since this year is the Seventieth Anniversary of Confederation, it seems to me appropriate that I should ask the Prime Minister of Canada to accept one of these medals."[3] Since 1927 there have been no additional presentations of the Confederation Medal, though the government continues to hold approximately twenty unissued examples.
The Confederation Medal was given new life in May 2008 when the first insignia of the Canadian Victoria Cross was unveiled. The Canadian VC contains bronze from not only the original VC cannon but also that of a bronze Confederation Medal. The symbolism of including metal from the first Canadian honour that bore the effigy of Queen Victoria was seen as a fitting way to further link the Canadian VC with Queen Victoria and the founding of the dominion.
The Confederation Medal
Silver Confederation Medal obverse.
Origins: Patterned after the various British coronation and commemorative table medals that find their origins in the sixteenth century.
Criteria: Awarded to senior Canadian and British dignitaries associated with Confederation, including the Fathers of Confederation, the first senators and MPs, the first members of the provincial legislative assemblies, various universities, and similar institutions.
Silver Confederation Medal reverse (Pinches re-strike, circa 1867).
Insignia: A circular medal 76 mm in diameter bearing on the obverse a veiled effigy of Queen Victoria circumscribed by the text VICTORIA D:G: BRITT: REG: F: D. The reverse bears an effigy of Britannia seated holding a scroll with the word CONFEDERATION on it with a lion at her feet. Attentively looking toward Britannia are four young maidens (representing the four original provinces), one each bearing a sickle, a paddle, a shovel, and an axe, the entire allegory circumscribed by the text JUVENTAS ET PATRIUS VIGOR / CANADA INSTAURATA 1867 ("Youth and patriotic strength, Canada inaugurated 1867").
Other: Issued in a maroon leatherette case usually stamped with the Wyon logo in gold foil on the inside lid. Medals were issued unnamed, though privately named examples are known. The design of the Confederation Medal was incorporated into a number of Indian Chief and Treaty Medals presented during the Victorian era. In honour of the centennial of Confederation, John Pinches and Company of London restruck the medal in sterling silver and gold. These restruck issues are numbered on the rim and are identifiable by the inclusion of the date 1967 in relief on the reverse.
Indian Chief Medal obverse, circa 1872.
Indian Chief Medal reverse, circa 1872.
Number: 551.
* • Queen Victoria: 1 gold.
* • Prince of Wales, governor general, Fathers of Confederation: 50 silver.
* • First members of the Dominion Parliament, first members of the provincial legislatures, other individuals and public institutions: 500.
# PART II
# 11
The Privy Councils: An Honour and High Office
In Canada titles such as "Right Honourable" and "Honourable" are often used in association with current and former holders of significant state offices and most notably membership in the Queen's Privy Council for Canada, what is colloquially known as the Privy Council. Prior to 1968, a number of Canadians were sworn in as members of the Privy Council of the United Kingdom. Prime ministers, eminent ministers of the Crown, and a number of Canadian high commissioners to London were all summoned by the Sovereign to be sworn as members of the Privy Council of the United Kingdom. Appointment was often used as an honour, since the designation was seen by many to eclipse that of a knighthood.
Members of the Privy Council in both the United Kingdom and Canada are sworn in before the Sovereign or her representative. The privy councillors are then entitled for life to be styled "Right Honourable" in the case of the United Kingdom and "Honourable" in the case of Canada. It also became a tradition in Canada for members to use the postnominials PC (CP in French) to distinguish them from other current or former office-holders entitled to the designation "Honourable." The postnominals PC are the highest in precedence in Canada immediately following the Victoria Cross and George Cross, and now the Cross of Valour, outranking all national orders and other decorations and designations. The letters PC are not formally used in the United Kingdom, since privy councillors are the only ones, besides peers, entitled to the designation "Right Honourable," though some members unofficially use the letters (interestingly, after all other letters in precedence) in order to distinguish themselves from peers. It is rare for the entire Privy Council to gather in either Canada or the United Kingdom. Such occasions are normally reserved for the accession of a new Sovereign and in connection with royal marriages.
The Imperial Privy Council
The Privy Council of the United Kingdom is also variously known as His/Her Majesty's Most Honourable Privy Council or, in the Canadian context, as the "Imperial Privy Council." The Privy Council in England dates back to A.D. 895 and began as a body of a dozen members charged with advising King Alfred. The council was reformed after the English Civil War, with a committee of the council, rather than the entire body, advising the Sovereign on matters of state and governance. It is largely from this tradition that our system of governance has evolved to the point that the Crown's chief minister and members of the Cabinet are members of the Privy Council. In Canada this means that members of the federal Cabinet are all members of the Queen's Privy Council for Canada, while members of the provincial and territorial cabinets are members of the various Executive Councils. It is worth noting that the Cabinet is in essence just a subcommittee of the broader Privy Council or Executive Council.
After Confederation it became the custom for a newly appointed governor general to be sworn into the Imperial Privy Council prior to departing for Canada — unless, of course, he was already a member of the council. Similarly, Canadian prime ministers were sworn into the Imperial Privy Council, with Lester B. Pearson being the last to hold such an appointment. Only three — Alexander Mackenzie, Sir John Abbott, and Sir Mackenzie Bowell — were never sworn into the Imperial Privy Council. Long-serving members of the federal Cabinet were also occasionally sworn into the Imperial Privy Council on the advice of the Canadian prime minister, and all except three chief justices of the Supreme Court of Canada were also sworn into the council from 1892 until 1968.[1] Chief justices were sworn in as a function of their involvement with the Judicial Committee of the Privy Council, which served as Canada's last court of appeal until 1949 when the Supreme Court was made the highest court in the land.
Number of Canadians Appointed to the Privy Council of the United Kingdom: 36.
The Queen's Privy Council for Canada
The Queen's Privy Council for Canada, more commonly known as the Privy Council, was established at the time of Confederation in 1867 as part of the Constitution Act, 1867. Members of the council swear an oath of membership and an oath of allegiance to the Queen, before the governor general, or in exceptional circumstances, before the Sovereign. They are thereafter entitled to carry the title "Honourable" for life and use the postnominal letters PC. Members have almost invariably been senators or MPs who serve as part of the ministry (Cabinet), though they remain privy councillors after their resignation from the Cabinet. During Paul Martin's brief period as prime minister, all parliamentary secretaries were also sworn in as members of the council, but this practice has not since been maintained. Similarly, since the establishment of the Security Intelligence Review Committee in 1984, members have been sworn into the Privy Council so that they can be privy to highly sensitive information. A number of members of the Royal Family have been sworn in as members of the Privy Council, notably Prince Edward, Prince of Wales (the future King Edward VIII); Prince Philip, Duke of Edinburgh; and Prince Charles, Prince of Wales. In the case of those royal members who go on to become Sovereign, their membership ceases upon accession to the throne.
In 1968, Prime Minister Lester Pearson requested that Canadian appointments to the Imperial Privy Council cease and that Her Majesty allow for the governor general, the prime minister, and the chief justice to be styled as "Right Honourable" by virtue of their offices. This meant that Canadians no longer had to be sworn into the Imperial Privy Council — a body of the Government of the United Kingdom — in order to be entitled to their traditional titles. The Queen agreed to allow for this change to the table of Canadian titles, and since that time the title "Right Honourable" has been granted to holders of the aforementioned high state offices.
In 1992, during the 125th anniversary celebrations of Confederation, a new practice was initiated to allow, on the advice of the prime minister, for certain eminent privy councillors to be designated as "Right Honourable" as an honour. This initial list of eight persons honoured in 1992 included such personalities as Ellen Fairclough, the first female federal Cabinet minister. Herb Gray, Canada's first Jewish federal Cabinet member and an MP for several decades, is the only other person to have been accorded this honour, which occurred in 2002.
Appointment to the Privy Council has occasionally been used as a significant honour. In 1967 during Canada's centennial year and again in 1982, in honour of the patriation of the Constitution, the serving provincial premiers were all sworn into the Privy Council. Beginning in 1891, speakers of the Senate and the House of Commons, and since 1979, governors general who were not previously members of the Privy Council, have usually been appointed in recognition of their services upon retirement. Commencing with Gordon Robertson in 1982, a number of long-serving clerks of the Privy Council have also been appointed to the council upon retirement.
In 1992 the prime minister advised the governor general to appoint various other eminent Canadians — who were not intended to serve as part of the ministry — to the Privy Council as a form of honour. At that time twenty-three citizens, including Maurice "Rocket" Richard, the Nobel Prize–winning physicist John Polyani, artist Alex Colville, and fifteen others were appointed to the Privy Council. The Queen was present for the swearing-in of a number of these individuals on Canada Day at Rideau Hall. To date only one person has been removed from the Privy Council — Conrad Black.
Number of Canadians Appointed to the Queen's Privy Council for Canada: 782.
Other Honourables
In addition to members of the Privy Council, a number of other state office-holders are also accorded the title "Honourable." Notable among this group are all current and former lieutenant governors, who have since 1927 been styled "Honourable" for life. Since 1867, members of the Senate of Canada are styled as "Honourable" for life, as well. The speaker of the House of Commons, speakers of the provincial and territorial legislatures, territorial commissioners, and territorial premiers are designated "Honourable" while in office. From time to time the governor general may also grant permission to retired speakers of the House of Commons and retired territorial commissioners to retain the designation "Honourable" for life.
Members of the provincial and territorial executive councils (cabinets) and speakers of the legislatures are styled as "Honourable" during their time in office. However, they cease to be entitled to use this designation upon relinquishing office. Thus, a former premier, former provincial minister of the Crown, or former speaker of a provincial legislature would not be styled "Honourable" unless they were sworn into the Queen's Privy Council for Canada.
Beginning in 2009, former members of the Executive Council in Nova Scotia are designated "Honourable" for life by a commission granted by the lieutenant governor under the Great Seal of the Province on the advice of the premier. In addition to being styled "Honourable" for life, former members of the Executive Council of Nova Scotia are also permitted to use the postnominals ECNS. Since 2014, all living former speakers have also been made honorary members of the Executive Council for Nova Scotia.
Since 1968 it has been the custom for members of Quebec's Executive Council not to use the designation "Honourable." Of historical interest is the fact that members of the provincial legislative councils — appointed chambers that existed in a number of provinces after Confederation — were styled as "Honourable" during their time in office. However, they did not retain the title after retirement — in large part because of the fact that most legislative councillors were appointed to office for life.
Members of the Bench
As noted above, the chief justice of the Supreme Court of Canada has been styled "Right Honourable" since 1968, though most previous chief justices had been sworn into the Imperial Privy Council and thus were also styled "Right Honourable." Puisne judges of the Supreme Court of Canada and judges of the Federal Court and Tax Court of Canada are styled "Honourable" while in office. Similarly, judges of the provincial territorial courts are styled "Honourable" while in office. Upon leaving office, certain judges may apply to the governor general on behalf of Her Majesty the Queen for permission to retain the title "Honourable." Permission is almost invariably granted to such requests. The table of titles to be used in Canada is administered and maintained by the Department of Canadian Heritage and was most recently updated in 1993.
# 12
The Order of Canada: The Basis for Canada's Honours System
The Origins of the Order of Canada
The idea of creating a Canadian order can be traced back to 1823 and Lord Bathurst, who was then Britain's colonial secretary. His proposal did not progress past the stage of a rough concept because King George IV did not favour the idea. This attempt was followed nearly fifty years later when Lord Monck, Canada's first governor general, made the same suggestion. Monck's proposal called for three levels: Knight Grand Cross of the Order of St. Lawrence (GCSL), Knight Commander of the Order of St. Lawrence (KCSL), and Companion of the Order of St. Lawrence (CSL). As with other imperial orders of chivalry, the top two levels would have conferred knighthood, and thus the recipient would have become "Sir Sean Morency, KCSL." It is likely that the order would also have carried an appellation such as "eminent," "exalted," or "excellent." One cannot help but wonder if Canada would still have discarded titular honours had a Canadian order of chivalry similar to that proposed by Monck been established shortly after Confederation. Alas, the Colonial Office was not willing to allow the new dominion to create such a potent symbol of national autonomy, and thus Canada continued to work within the existing imperial honours system. Indeed, the Order of St. Michael and St. George was altered in its purpose to meet the demands of Canada and other self-governing parts of the British Empire, and thus became in many ways the premier honour for a Canadian to receive.
The origins of the Order of Canada as we know it today can clearly be traced back to Vincent Massey, whose persistence and ideas played such a central role in the creation of a uniquely Canadian honour that he initially styled the Order of St. Lawrence. The five tiers of Massey's order were: Grand Commander of the Order of St. Lawrence (GCSL), Grand Officer of the Order of St. Lawrence (GOSL), Companion of the Order of St. Lawrence (CSL), Officer of the Order of St. Lawrence (OSL), and Member of the Order of St. Lawrence (MSL). Massey first broached the subject in 1935 with newly installed Governor General Lord Tweedsmuir, who was very keen on encouraging the creation of Canadian institutions and subsequently brought the subject up with Prime Minister William Lyon Mackenzie King. Mackenzie King was not enthusiastic about the prospect of reintroducing honours in Canada, let alone creating an indigenous award. Canada's longest-serving prime minister had a phobia when it came to honours, going as far as to ask the British Central Chancery of the Orders of Knighthood to cancel his appointment as a Companion of the Order of St. Michael and St. George more than thirty years after he received it. This did not prevent him in later life from accepting the Commonwealth's premier non-titular award when he was appointed to the Order of Merit in 1947 and a number of foreign awards.
Royal Order of Canada diagram.
Massey initially proposed a Canadian order called the Order of St. Lawrence, consisting of two levels: Companion and Officer. It was a concept he reintroduced again without success during the early stages of the Second World War. Throughout the war a special committee made up of senior civil servants, the Awards Coordination Committee (ACC), met regularly to discuss the issue of honours in Canada. While the committee's task was primarily to develop policy with relation to British awards being bestowed upon Canadians, it would often delve into the subject of creating a Canadian order. To this end, between 1941 and 1946, it submitted no fewer than five different proposals to the prime minister. Each was in turn rejected. In 1942 a special parliamentary committee, the Special Committee on Honours and Awards, called for the creation of a Canadian order, a recommendation that was also ignored. It was at this time that the actual name "Order of Canada" originated as part of Massey's proposal for a Royal Order of Canada — though it was the undersecretary of state, Ephriam Coleman, and Major-General Harry Letson who simplified the name to the Order of Canada. This name persisted in proposals until the creation of the Order of Canada in 1967, though other titles, ranging from the Order of the Beaver, to the Order of St. Lawrence, to the Royal Elizabethan Order, were often suggested.
Despite the failure of its various proposals, the ACC did manage to have the Canada Medal established in 1943. It was a curious award, to say the least, since it was to be awarded to everyone from privates to heads of state. While extensive lists of recipients were composed, the medal was never awarded. Mackenzie King wanted nothing to do with honours, especially the Canada Medal; in his mind the existing British honours system, though flawed, worked well in Canada, and he had no desire to make forays into the honours debate by creating a Canadian system.
After the end of the Second World War, the demand for civilian honours tapered off and the topic of creating a Canadian order was once again put aside. Lord Alexander, Canada's governor general from 1946 to 1952, called for the creation of an Order of Canada in 1948, but this, too, failed, though he did come up with the ribbon design used by the Order of Canada today. The subject was again reintroduced in a secret report of the Royal Commission on the Arts, Letters, and Sciences, which was, not coincidentally, chaired by Vincent Massey. This time Massey proposed a five-level Order of St. Lawrence based on a structure similar to the Order of the British Empire and France's Légion d'honneur. Most significantly, Massey proposed that the honours list for this new order be composed by a non-partisan committee, quite unlike the practice that had previously been followed in Canada and Britain where the prime minister had a direct influence over the flow of honours. Although the proposal was given to the leaders of the opposition parties, it failed to gain the approval of then Prime Minister Louis St. Laurent, who was in part concerned about sanctioning the creation of an order that bore a name so similar to his own. One can only imagine the reaction of political opponents if St. Laurent were to bestow an order bearing his name to loyal members of the Liberal Party!
Throughout the late 1950s and early 1960s, the Department of National Defence developed a variety of proposals for the creation of an Order of Canada. These were based upon the Massey Commission and Alexander proposals, though none met with success.
The adoption of the Maple Leaf Flag and the establishment of the Order of Canada were really part of a larger project on the part of the government to endow Canada with the requisite symbols of nationhood. Before Lester Pearson became prime minister in 1963, he had supported the adoption of a new national flag to replace the Canadian Red Ensign. His interest in creating a Canadian order was piqued by his parliamentary secretary, John Matheson, who shared his interest in both the flag and the order. Pearson was most concerned with the flag issue, and thus the question of establishing a Canadian order was temporarily sidetracked. More than a year after the adoption of the Maple Leaf Flag — with tensions still high — Pearson and Matheson discussed the possible structure of a Canadian order. Pearson hoped to have an order established in time for the Centennial of Confederation in July 1967. This left less than fifteen months in which to gain the approval of the Cabinet and the Queen, to have the insignia designed and manufactured, and most importantly, to draw up a list of worthy recipients.
Order of Canada prototype sketch.
In March 1966, Matheson was dispatched to Vincent Massey's country estate, Batterwood, to discuss the subject. Their deliberations were productive, and upon his return to Ottawa, Matheson presented Pearson with a plan for a three-tiered order consisting of Companions, Officers, and Associates. Massey still favoured the name the Order of St. Lawrence, while Pearson and Matheson wanted the new honour to be called the Order of Canada. Over the next six months various changes were made to the proposal, and the constitution was developed. The yet unnamed order was to consist of three levels: Companion, Member, and Associate.
The proposal for the establishment of the Order of Canada was presented to Cabinet in November 1966, and the reception was frosty. There was opposition to the order from a number of ministers: some viewed it as elitist and not reflective of the sort of honours system Canada needed; some went as far as to question the need for an honours system at all. Pearson yielded and agreed that the Order of Canada would consist of one level only, and Cabinet finally agreed to allow the plan to go forward.
Pearson was concerned that a single-level order would not reward all those who deserved recognition; after all, his primary motive for proposing the order was to see those who contributed at the local level recognized. Two additions were subsequently made: the Medal of Service, which was in essence a junior level of the Order of Canada; and the Medal of Courage, a separate bravery award that, though part of the order, was quite an independent entity. The Medal of Service and Medal of Courage were approved by Cabinet on the basis that they were not really part of the order, but this would prove a mistake: Pearson's assistants — Gordon Robertson, Jack Hodgson, and Michael Pitfield — had cleverly described the medals to seem like minor additions when, in fact, they were integral to the structure of the order. The Queen approved the creation of the Order of Canada on March 21, 1967. The Order was announced on April 17 and public reaction was favourable, though there were some initial concerns that the order would become a political tool or patronage plum.
The first honours list was published in the Canada Gazette and released to the public on July 7, 1967. The list included such notables as Dr. Wilder Penfield, Vincent Massey, Madame Pauline Vanier, Major-General George Pearkes, M.J. Coldwell, Gaetan Gélinas, Maurice Richard, and Marlene Streit. In total, thirty-five Companions and fifty-five Medals of Service were appointed that month. A second list of fifteen Companions and forty-five Medals of Service was issued in December 1967. The first recipient of the Order of Canada was Roland Michener, who was invested by the Queen on July 6, 1967, at Government House.
Almost immediately after the first appointments were made in 1967, issues arose with the structure of the order. No awards of the Medal of Courage had been made, so it did not pose an immediate problem, but the Medal of Service was quite another issue. It was regarded by many as a second prize to the Companion level, and this was not only because more Medals of Service were awarded than Companions, but also because the design of the insignia implied inferiority. The Companion's insignia was quite beautiful in gold and enamel, while the Medal of Service claimed a rather plain and smaller — though not unattractive — insignia. There was the additional problem that the Medal of Service was intended to recognize important national service, yet there was a desperate need for an award to recognize outstanding local service of a significant nature.
By 1968 the Advisory Council of the Order of Canada had decided that the order should be restructured into three levels: Companion, Officer, and Member — the same arrangement that Pearson had proposed to the Cabinet in 1966 when it was deemed unacceptable. This time, however, there was little difficulty in securing acceptance. Those who received the Medal of Service were permitted to exchange that medal for an Officer's insignia, and the Medal of Service was cancelled.
The restructured Order of Canada came into being in 1972. Since that time there have been few structural changes, aside from enlargements and an alteration in the mechanism through which honorary members are appointed. In 2013 a third division was added to the order — that of extraordinary membership.
Appointments to the Order of Canada are made by the governor general on behalf of the Queen. The governor general receives the honours list from the order's Advisory Council, which is chaired by the chief justice of the Supreme Court of Canada and which includes the secretary to the governor general (who is also secretary general of the order), the deputy minister of Canadian Heritage, the clerk of the Privy Council, the president of the Royal Society of Canada, the president of the Association of Universities and Colleges of Canada, a representative from the protective services, a scientific (non-medical) member, a charity/religious representative, and up to five other members representing the regions: Atlantic, Quebec, Ontario, Prairies, and the West. Nominations are received from the general public, and the entire process is insulated from political influence.
Although the Order of Canada's 1967 Constitution provided for non-Canadian citizens to be appointed as honorary members on the advice of the Cabinet, it was not until 1981 that the first and only such appointment was made. Zena Sheardown, along with her husband and other members of the Canadian embassy's staff in Tehran, sheltered six Americans during the Islamic Revolution in Iran in 1979. Mrs. Sheardown's role was particularly central, since four of the six "house guests" were billeted with her. She was also at the greatest risk, since she was not a Canadian citizen and was therefore not covered by diplomatic immunity. In 1981 she became an honorary Member of the Order of Canada; when she later became a Canadian citizen, her honorary membership was cancelled and she was simultaneously appointed a Member of the Order of Canada, general division.
When the order was established, it consisted of two divisions: a general division for Canadian citizens and an honorary division, which is open to Commonwealth and foreign citizens. Unlike many other countries, however, Canada does not engage in the diplomatic game of exchanging honours as gifts.
The addition in 2013 of an extraordinary division to the order was made to allow for the appointment of members of the Canadian Royal Family to the order, and also to correct a long-standing anomaly whereby the governor general and the vice-regal spouse were appointed to the general division of the order without deducting their appointments from the numerical limit on membership included in the Constitution of the order. In 2013 all living governors general and vice-regal spouses were transferred to the extraordinary division of the order.
As with every Canadian honour, membership in the Order of Canada can be revoked and this has occurred on a number of occasions. The first happened when Zena Sheardown became a Canadian citizen and moved from being an Honorary Member to a regular Member of the order, though this was more of a transfer from one division to another. The ignominious terminations include Alan Eagleson (1998), David Ahenakew (2005), T. Sher Singh (2008), Steve Fonyo (2010), Garth Drabinsky (2012), and Conrad Black (2013). There have also been nine resignations from the order.
To date, more than six thousand Canadians have been appointed to the Order of Canada. It is a national honour open to all fields and recognizes local as well as international contributions. Every recipient of the Order of Canada also receives a certificate bearing his or her name, the signature of the governor general and secretary general of the order, and the seal of the order.
When the Order of Canada was created in 1967, special allowances were made to permit it to be awarded to non-Canadians. These honorary awards were to be made on the advice of the Queen's Privy Council for Canada (the federal Cabinet). For various reasons only one award was made via this method between 1967 and 1998.
In 1985–86, several proposals were introduced to see a wholly separate Canadian order established with membership limited to non-Canadians only. The Mulroney government briefly flirted with the idea of engaging in the diplomatic honours exchange game, whereby Canadian high commissioners and ambassadors would be permitted to accept foreign decorations merely for serving in that country, while foreign diplomats and visiting heads of state would be presented with the new Canadian order. Designs and a ribbon were drawn up by Bruce Beatty. The ribbon was white-red-white, the opposite of the Order of Canada's ribbon. The insignia was to follow the same idea, and the white enamel on the snowflake was to be replaced by red enamel. At some point consideration was given to eliminating the honorary division of the Order of Canada and creating an entirely new Order of Merit for foreigners only. By 1987 the entire proposal was shelved.
Enlargements and Changes to the Order of Canada, 1967–2014 | Companion | Officer | Member
---|---|---|---
1967 | 50 annual (150 max.) | 100 (Medal of Service) | None
1972 | 15 annual (150 max.) | 40 | 80 (new level)
1983 | 15 annual (150 max.) | 46 | 92
1994 | 15 annual (150 max.) | 50 | 100
1995 | 15 annual (165 max.) | 50 | 100
1997 | Change in policy toward honorary appointments
1998 | 15 annual (165 max.) | 52 | 106
1999 | 15 annual (165 max.) | 64 | 136
2000 | Expansion of the Advisory Council by five members
2013 | Addition of the Extraordinary Division of the order
Sovereign of the Order of Canada
Sovereign of the Order of Canada insignia.
Origins. Every national order has a titular head. In the case of countries with monarchical systems of government, it is generally the Sovereign; in the case of republics, it is the president. As Canada is a constitutional monarchy, it is natural that the Queen serve as the Sovereign of the Order of Canada — it is, after all, from the Crown that all official honours flow.
It has long been a tradition that the Sovereign of an order possess a special insignia to denote his or her position. These insignia are usually encrusted with precious stones and different in some way from those awarded to the general membership of the order. The Order of Canada follows this tradition, and in 1967 it was decided that a special Sovereign's insignia should be commissioned and presented to the Queen. The design was completed in 1968, and the insignia was presented to Queen Elizabeth II by Governor General Roland Michener on June 23, 1970, at Buckingham Palace.
Queen Elizabeth II and Roland Michener, July 1967, shortly after the Queen invested Michener as the first Companion of the Order of Canada.
Insignia: A six-armed cross of 18-carat gold enamelled white with a large square diamond set between each of the arms. In the centre there is a maple leaf surrounded by an annulus, both set with calibre-rubies pavé. On the annulus is the motto in pierced gold. Above the cross is a gold Royal Crown with the cap of maintenance enamelled red and the ermine enamelled white. The arches are set with twenty-one diamonds, with a larger one in the orb. The base is set with a sapphire, two emeralds, and two rubies. The reverse is plain, aside from CANADA in gold. The main difference from the other insignia of the order is that the crown is above the snowflake rather than resting on the annulus.
Other: The Sovereign's insignia was manufactured by Garrard & Co.
Chancellor of the Order of Canada
Chancellor of the Order of Canada insignia.
Origins: While every order has a titular head, tradition has also called for there to be a senior official known as either a grand master or a chancellor. Prime Minister Lester Pearson decided that the term grand master was a little grandiose for Canada, thus chose the title of chancellor. Given that the governor general is the Queen's representative in Canada, it was decided that he or she should also serve as the Chancellor of the Order and perform most of the ceremonial and administrative duties relating to certain aspects of the order. The Chancellor of the Order of Canada is also styled the "Principal Companion of the Order of Canada." When a governor general's term expires, he or she ceases to be Chancellor and Principal Companion of the Order of Canada but remains an extraordinary Companion of the Order.
Insignia: The Chancellor's Chain was produced by the Royal Canadian Mint. Like the other insignia of the order, it was designed by Bruce Beatty. Marvin Cook and Argo Aarand made the various parts of the chain from gold, and Aarand enamelled them in a small kiln in the basement of his home. The chain is made of twenty-three devices linked by a double row of small gold links. Twelve of these devices are miniature replicas of the white snowflake; alternating with these are ten devices, each in the form of a red maple leaf on a white background encircled by the red annulus bearing the motto of the order. The chain is completed by a centre device in the form of the shield from the Arms of Canada ensigned by the Royal Crown, each in their proper colours. A Companion's insignia is hung from this device. The insignia of the Order of Canada that hangs from the Chancellor's Chain was manufactured by Garrard & Co. The reverse is the same as that of a regular Companion's insignia, though a small box that usually contains a three-digit number bears the legend -C-.
The chain — made over a nine-month period during which Aarand and Cook devoted their lunch hours to the project — was presented to Governor General Michener by E.F. Brown, the acting master of the Royal Canadian Mint, on December 22, 1968.
Order of Canada certificate.
Other: The Companion's insignia is detachable from the Chancellor's Chain, though this is rarely done. The Chancellor's Chain is worn primarily on special occasions such as investitures and sometimes the opening of Parliament. The chain is housed in a large red leatherette case whose lid is embossed with the insignia of the order in gold.
Number: 9.
* • Right Honourable Roland Michener, PC, CC, CMM, OOnt, CD, QC: 1967–74.
* • Right Honourable Jules Léger, PC, CC, CMM, CD: 1974–79.
* • Right Honourable Edward Schreyer, PC, CC, CMM, OM, CD: 1979–84.
* • Right Honourable Jeanne Sauvé, PC, CC, CMM, CD: 1984–90.
* • Right Honourable Ray Hnatyshyn, PC, CC, CMM, CD, QC: 1990–95.
* • Right Honourable Roméo LeBlanc, PC, CC, CMM, ONB, CD: 1995–99.
* • Right Honourable Adrienne Clarkson, PC, CC, CMM, COM, CD: 1999–2005.
* • Right Honourable Michaëlle Jean, PC, CC, CMM, COM, CD: 2005–10.
* • Right Honourable David Johnston, CC, CMM, COM, CD: 2010 to present.
Companion of the Order of Canada
Companion of Order of Canada obverse.
Early Companion of the Order of Canada obverse.
Origins: The Companion level of the Order of Canada is the highest, the oldest, and understandably, the most prized part of the order. The first person invested with the order was Roland Michener, who was presented with the very first insignia by the Queen at Government House on July 6, 1967. The youngest Companion of the Order of Canada was Terry Fox, who was twenty-two at the time of his appointment.
Originally, the number of living companions was limited to 150, not including ex officio members. In 1995 this was increased to 165. From 1967 until 1995, a maximum of five Companions could be appointed each year, though a provision was made for fifty Companions to be appointed in the Centennial year — there was, after all, a significant backlog of worthy recipients, there having been no civilian honours list since 1946. Since 1995 the annual maximum has been fifteen.
Criteria: For outstanding achievement and merit of the highest degree, especially service to Canada or to humanity at large.
Insignia: A gold stylized snowflake measuring 57 mm across covered in white enamel. The obverse bears a single red maple leaf in the centre surrounded by a red circlet containing the words DESIDERANTES MELIOREM PATRIAM ("They desire a better country") in gold. The circlet is surmounted by a Royal Crown, which on the insignia of all Companions is set with small precious stones. The reverse is also enamelled in white with CANADA in the centre. Below it a small rectangular box contains a raised issue number.
Companion of the Order of Canada reverse.
Companion lapel pin.
Suspension: A gold ball is mounted to the apex of the uppermost arm of the snowflake and through this passes a small gold ring, through which passes a 30 mm straight hanger and the ribbon. The Companion's insignia is worn around the neck, though women have the option of wearing it on their left shoulder on a bow.
Ribbon: A white ribbon 38 mm wide edged with 9 mm of red. Recipients wear a red maple leaf on their undress ribbon.
Lapel Badge: A small gold snowflake enamelled white bearing a maple leaf in its centre in red enamel.
Postnominals: CC.
Number of Appointments: 475.
Other: In mid-1984 the four small dots located in the circlet that bears the motto of the Order of Canada were replaced with a small quadrangle because the enamel around the dots tended to flake off. In 2001 the thickness of the suspender was changed from 1.25 mm to 1.75 mm.
Medal of Courage of the Order of Canada
Origins: The Medal of Courage was not part of the ori-ginal Order of Canada proposal submitted to the Cabinet in November 1966. It was added a month later and was announced as part of the Order of Canada when the new honours system was unveiled in April 1967. However, the first Order of Canada list only contained appointments of Companions and the awarding of Medals of Service. The Government Decorations Committee, the body charged with administering the Medal of Courage, did not put forward any names, since Prime Minister Pearson wished for the first recipient of the Medal of Courage to be presented with the medal at a special event and not be lumped in with the other recipients of the Order of Canada. There was to be no limit on the number of awards made annually.
Originally, it was to be known as the Gallantry Medal of the Order of Canada, but the name was changed to Medal of Courage because this translates into French more easily. The government planned to continue using the Victoria Cross, George Cross, and Queen's Commendation for Bravery as the principal bravery awards, while the Medal of Courage was to take the place of the George Medal. The criteria for the last two medals were very similar.
Medal of Courage of the Order of Canada obverse.
Medal of Courage of the Order of Canada reverse.
Placing a gallantry award within an order was not unique, and it closely followed the pattern established by the Order of the British Empire, which included the Empire Gallantry Medal, an award that had itself been replaced by the George Cross in 1940. Despite this change, the Order of the British Empire continued to be awarded for gallantry until 1974, and the last Canadian gallantry appointment to that order was made in 1968.
In March 1968, Pearson decided that Canadians should no longer be eligible for British bravery awards such as the Victoria Cross, the George Cross, and the Queen's Commendation for Bravery. Suddenly, the Medal of Courage was to be elevated from an intermediate bravery award to one that would recognize all acts of bravery. This idea was almost immediately dismissed as impractical by officials at Government House, and a new proposal emerged to have the Medal of Courage divided into three levels. This, too, was rejected, and by October 1968 a new proposal arose for the establishment of three separate bravery awards, but it was not until 1972 that the final proposal was adopted. The Medal of Courage, having never been awarded, was cancelled and replaced by the Cross of Valour, the Star of Courage, and the Medal of Bravery.
Criteria: Awarded to a person who performs an act of conspicuous courage in circumstances of great danger.
Insignia: A gold stylized snowflake measuring 34 mm across, the obverse depicting a single maple leaf enclosed within two circles and surmounted by a Royal Crown. The reverse bears a single silver disc with the word COURAGE around the top of the disc with four small dots in its base.
Suspension: A small hollow silver ball at the apex of the uppermost arm of the snowflake through which passes a ring and the ribbon.
Ribbon: A 34 mm white ribbon edged with 7 mm of red. A recipient was to wear a gold maple leaf in the centre of the undress ribbon. The ribbon used for female recipients was 32 mm wide.
Bars: Subsequent awards were to be recognized with a single gold bar bearing a maple leaf in the centre. No bars were manufactured.
Lapel Badge: None was ever issued.
Postnominals: CM.
Other: The insignia was made of gilded sterling silver and was manufactured by Garrard & Co. Twenty-four Medals of Courage were made (twenty men's issue and four women's issue on bows). Each was contained within individual black leatherette cases stamped with C.M. On February 23, 1984, the Royal Canadian Mint melted eighteen of the Medals of Courage, leaving only six. Four of these are held by the Government of Canada, while two are in private hands.
Number: None.
Medal of Service of the Order of Canada
Medal of Service of the Order of Canada obverse.
Medal of Service of the Order of Canada reverse.
Origins: As with the Medal of Courage, the Medal of Service was not included in the early proposals made to the Cabinet for the Order of Canada. The Cabinet, as previously noted, turned down Prime Minister Pearson's original proposal for a three-tiered Order of Canada on the grounds that it was elitist. As a partial remedy, the Medal of Service was developed to serve as a junior version of the Companion level. It was added to the proposal for the Order of Canada in February 1967 and was announced in April 1967 alongside the Companion of the Order of Canada and the Medal of Courage. Although part of the Order of Canada, it was called the Medal of Service so that the Cabinet would approve it without seeing it as a different level of the same order. A limit of fifty Medals of Service could be awarded each year, with the exception of the Centennial year of 1967 in which provision was made for one hundred Medals of Service to be awarded. There was no limit on the overall number of holders of the Medal of Service of the Order of Canada.
Almost immediately there were problems with the Medal of Service, since it was seen as inferior to the Companion level. In a few cases people actually refused to accept the Medal of Service because they felt entitled to be appointed as a Companion. In October 1968, at about the same time that a replacement for the Medal of Courage was being considered, the government and the Advisory Council of the Order of Canada tentatively agreed that the Medal of Service should be cancelled and replaced by two new levels: Officer and Member. The restructuring of the Order of Canada was completed in 1972, and recipients of the Medal of Service were invited to exchange their SM insignia for that of an Officer of the Order of Canada.
Criteria: Awarded for achievement and merit of a high degree, especially service to Canada or to humanity at large.
Insignia: A silver stylized snowflake measuring 34 mm across, the obverse depicting a single maple leaf enclosed within a circle and surmounted by a Royal Crown. The reverse bears a single silver disc with the word SERVICE around the top with four small dots in its base.
Ribbon: A 34 mm white ribbon edged with 7 mm of red. A recipient was to wear a silver maple leaf in the centre of the undress ribbon. The ribbon used for female recipients was 32 mm wide.
Suspension: A small hollow silver ball at the apex of the uppermost arm of the snowflake through which passes a ring and the ribbon.
Bars: No bars could be awarded to the Medal of Service, though one could be elevated to the level of Companion. This occurred in three instances only.
Lapel Badge: None was ever issued.
Naming: The reverse lower arm of the medal was engraved with the name of the recipient in small capitals. Medals awarded to men simply had their initials and surname (V.F. MCADAM), while those awarded to women bore both their given name and surname (ELSINORE BURNS). A number of unnamed examples are known to exist.
Postnominals: SM.
Other: The insignia was made of sterling silver and manufactured by Garrard & Co. While 319 Medals of Service were approved by the governor general, only 294 were actually awarded. From 1967 to 1971, Garrard delivered a total of 300 Medals of Service to the Canadian government. On February 23, 1984, the Royal Canadian Mint melted 189 of the exchanged Medals of Service, leaving fewer than 120 in existence.
Number: 319 appointments, 294 actual engraved awards.
Officer of the Order of Canada
Officer of the Order of Canada Garrard issue reverse.
Officer of the Order of Canada obverse.
Origins: This level of the Order of Canada, instituted in March 1972 and made effective for July 1, 1972, is the successor to the Medal of Service. The award serves as recognition for significant lifetime contributions to Canada or to the international community. A maximum of seventy-two persons can be appointed each year. There is no limit on the overall membership.
Criteria: For achievement and merit of a high degree, especially service to Canada or to humanity at large.
Insignia: A gold stylized snowflake measuring 46 mm across covered in white enamel. The obverse bears a single gold maple leaf in the centre surrounded by a red circlet that contains the words DESIDERANTES MELIOREM PATRIAM in gold. The circlet is surmounted by a Royal Crown. The reverse is plain brushed gold (silver gilt) with a raised disc bearing CANADA with a three- or four-digit number engraved below.
Suspension: A gold ball is mounted to the apex of the uppermost arm of the snowflake and through this passes a small gold ring, through which passes a 20 mm straight hanger and the ribbon. The Officer's insignia is worn around the neck, though women have the option of wearing it on their left shoulder on a bow.
Ribbon: A white ribbon 38 mm wide edged with 9 mm of red. A recipient wears a gold maple leaf on the undress ribbon.
Lapel Badge: A small gold snowflake enamelled white bearing a gold maple leaf in its centre.
Postnominals: OC.
Officer lapel pin.
Officer of the Order of Canada Rideau issue reverse.
Other: The maker's mark is impressed on the lower reverse arm of the insignia.
Number of Appointments: 2,083.
Member of the Order of Canada
Member of the Order of Canada obverse.
Member of the Order of Canada Garrard issue reverse.
Origins: This level was added after the 1972 restructuring of the Order of Canada. In many ways it is the most important and most inclusive level of the order, since it recognizes services of a local or very specific nature. Prime Minister Pearson intended the Order of Canada to include people who had made contributions to Canada not only at the national level but also at the most local level. When the Advisory Council of the Order of Canada began to review the order's structure in 1968, it was apparent that those who had made significant contributions at the local level or in a specialized field were not eligible to be appointed as Companions of the Order, nor did they qualify to be awarded the Medal of Service, since both required a contribution of a national or international calibre. Fewer than 10 percent of those who had made local contributions were awarded the Medal of Service. A maximum of 136 persons can be appointed each year. There is no limit on the overall membership.
Criteria: For distinguished service in or to a particular locality, group, or field of activity.
Insignia: A silver stylized snowflake measuring 38 mm across covered in white enamel. The obverse bears a single silver maple leaf in the centre surrounded by a red circlet that contains the words DESIDERANTES MELIOREM PATRIAM in silver. The circlet is surmounted by a Royal Crown. The reverse is plain brushed silver with a raised disc bearing CANADA with a three- or four-digit number engraved below.
Member of the Order of Canada Birks issue reverse.
Member lapel pin.
Suspension: A silver ball is mounted at the apex of the uppermost arm of the snowflake and through this passes a silver ring and the ribbon. The member's insignia is worn on the left breast, though women have the option of wearing it on their left shoulder on a bow if they have no other medals.
Ribbon: A white ribbon 38 mm wide edged with 9 mm of red. A recipient wears a silver maple leaf on the undress ribbon.
Postnominals: CM.
Other: A maker's mark is impressed on the lower reverse arm of the insignia.
Number of Appointments: 3,890.
Insignia Manufacture: The following section outlines the various firms that have produced the insignia of the Order of Canada, along with details related to dates of manufacture and placement of makers' marks. The dates listed are the year of manufacture; in some years additional stock was ordered by the Chancellery, resulting in gaps in production years when additional stock was not required for investitures. All insignia produced until 2014 were finished with vitreous enamels, while all insignia produced by the Royal Canadian Mint have a fill like hard epoxy.
Set of Royal Canadian Mint issue Order of Canada insignia.
Order of Canada Insignia Manufacturers and Material
Manufacturer | Companion | Officer and Member
---|---|---
Garrard & Co. | Marks: Un-hallmarked
Location: Reverse of the hanger (rare)
Material: 18-carat gold
Dates: 1967–82 | Marks: G&CO and usual British hallmarks
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OC)
Other: Some early OC issues carry no hallmarks; in 1978 a small number were marked SILVER in place of G&CO
Dates: 1972–82/3
Rideau Ltée | Marks: Early insignia unmarked; post-2004 insignia are laser-engraved with RIDEAU + R STER
Location: Hanger or reverse of mounting ball or the edge of the lowest arm
Material: 18-carat gold from 1984–96; thereafter .925 silver gold-plated
Dates: 1985–96, 2004–05 | Marks: RIDEAU + R STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OC)
Dates: 1982/3–96, 2004–05
Birks
(Pressed Metal Products) | Marks: BIRKS STER
Location: Reverse of the hanger
Material: .925 silver gold-plated
Dates: 1998–2006 | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OC)
Dates: 1998–2007
Pressed Metal Products | Marks: PMP STER
Location: Reverse of the hanger
Material: .925 silver gold-plated
Dates: 2011 | Marks: PMP STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OC)
Dates: 2010–13
Royal Canadian Mint | Marks: STERL on reverse of hanger
Location: Reverse of the hanger
Material: .999 silver (gold-plated for OC)
Dates: 2014– | Marks: RCM logo
Location: Reverse of the lower arm
Material: .999 silver (gold-plated for OC)
Dates: 2014–
# 13
The Order of Military Merit
Orders are a product of the Middle Ages when they were developed to recognize defenders of the realm. To be successful in the civil sphere, one was invariably required to be proficient in the art of war; thus, the concept of honours in general finds its root in military service. The best known of these military orders is Britain's Order of the Bath. Over time, however, the Order of the Bath was split into civil and military divisions, and Canadians would go on to be recognized with it — most notably Sir John A. Macdonald, who was made a Knight Commander of the Order of the Bath in 1867 and was later elevated to a Knight Grand Cross of the Order of the Bath. The military division of the Order of the Bath includes some of Canada's most prominent military leaders. Sir Arthur Currie, Sir William Otter, and Sir Richard Turner were all appointed Knight Commanders prior to or during the First World War.
Other military orders of antiquity include France's Royal Military Order of St. Louis. Founded by Louis XIV on April 5, 1693, it was awarded to those of the Catholic faith who had rendered distinguished services to the Crown and to France. It was the principal honour bestowed on members of the Canadian Militia and French Army serving in Canada prior to 1759.
France's Ordre du Mérite militaire.
It is primarily in these two orders that we find the basis for the structure and ethos of the Order of Military Merit. Both consisted of three levels and were awarded for exceptionally meritorious service; both also have a direct connection to Canadian history. Although just over forty years old, the Order of Military Merit can claim a much more ancient heritage through these ancestors. Within these traditional orders, military rank determined which level one could be appointed to. Generals were usually appointed at the Knight Grand Cross level, while majors were made Companions or Chevaliers.
It is hard to believe that for nearly thirty years between the end of the Second World War and the establishment of the Order of Military Merit in 1972, members of what are today the Canadian Armed Forces were not eligible for any honours or awards other than for bravery or long service. This vacuum of honours for meritorious military service was one of the main catalysts that prompted the creation of the Order of Canada, and the role of the Canadian Armed Forces in pushing for a uniquely Canadian honours system should not be underestimated.
With the foundation of the Order of Canada in 1967, one would assume that the military's desire for a system of honours and awards had been met. This was, however, not the case. Early proposals for the order dating back to 1944 had suggested that it include civilian and military divisions, a concept derived from two of the main imperial orders of chivalry — the Order of the Bath and the Order of the British Empire. These recommendations were not followed. Also, in 1966, the Department of National Defence (DND) had developed proposals for a Forces Meritorious Cross and a Forces Meritorious Medal, but these plans were shelved in favour of the Order of Canada, which was viewed as the best solution to the absence of a system of military exemplary service awards. The general feeling within DND was that it would therefore be used to recognize both civil and military achievements in a fashion similar to France's Légion d'honneur. As it turned out, in its early days the Order of Canada was being conferred almost exclusively as a civilian honour, and those military officers who were admitted to the new order were either retired or being recognized for non-military accomplishments.
Grand Cross of the Order of Military Merit prototype.
Within six months of the first investiture ceremony in 1967, changes to the order's structure were being sought. The body that was charged with making improvements — the Working Group on Honours and Awards — briefly considered the possibility of dividing the Order of Canada into military and civil divisions, though this idea was not well received by the order's Advisory Council.[1] There was consensus that a separate award for the military was needed because "recognition of distinguished service has important moral implications."[2]
As a result of a submission to the Cabinet seeking approval for the award of the Order of the British Empire (three MBEs and two BEMs) to Canadians for service in United Nations missions, the Cabinet agreed in principle on May 5, 1966, to an examination of the desirability of establishing Canadian awards for meritorious military service. The government's Decorations Committee considered a related proposal in early December 1969, and a memorandum was sent to the Cabinet later that month. The Cabinet responded by requesting a more detailed proposal. The solution emerged in a report of the Decorations Committee issued in January 1970. That report proposed the creation of a new order consisting of three levels to recognize "Military Merit."[3] By December 1971 the idea had evolved into the Order of Military Merit, which was to consist of three levels: Commander, Officer, and Member.
Grand Officer of the Order of Military Merit prototype.
An earlier proposal for the order to consist of five levels was rejected, since the order would not have a structure similar to the soon-to-be-reorganized Order of Canada, which would come to have three levels. Nevertheless, designs were drawn up by Bruce Beatty for a five-level order, which was to include the designations Grand Commander, Grand Officer, Commander, Officer, and Member. There was also to be a Medal of the Order of Military Merit. This version of the Order of Military Merit featured an effigy of Queen Elizabeth II rather than a maple leaf and crown in the centre. Chief of the Defence Staff General Jacques Dextraze mounted an attempt to reintroduce an earlier proposal that called for a five-level structure in 1973; however, this was politely shelved.
Recipients of the new order were to be selected by a special council composed of one member appointed by the governor general and five others who are members of the Canadian Armed Forces and are appointed by the chief of defence staff. There are strict limits on the total number of annual awards, capped at one-tenth of 1 percent of the Canadian Armed Forces' strength. Of these awards only 5 percent can be made at the Commander level, 20 percent at the Officer level, and the remaining 75 percent at the Member level.
The motto of the order is Officium Ante Commodum (roughly translated as "Service before self"). The name of the order can be traced back to the Kingdom of France where the Ordre du Mérite militaire was awarded to Protestant military officers who had rendered outstanding service to the Crown.[4] As is the case with the Order of Canada, the Queen is Sovereign of the Order of Military Merit and wears a special jewelled Sovereign's insignia. The governor general serves as Chancellor of the Order of Military Merit during his or her time as governor general and commander-in-chief of Canada, while the chief of the defence staff (CDS) is the Principal Commander while in office.
The order consists of three divisions: the general division, for members of the Canadian Armed Forces; the honorary division, which is open to members of allied foreign armed forces; and the extraordinary division, which is open to members of the Canadian Royal Family and serving and retired governors general. The extraordinary division was established in 2013, and the first person to be honoured with a direct appointment was His Royal Highness The Duke of Edinburgh, who was appointed as a Commander of the Order of Military Merit on April 23, 2013, and invested on April 26. One non-Canadian has been appointed an honorary member of the order: General Richard Myers of the United States Air Force was appointed as an honorary CMM in February 2005. To date there has also been one termination and two resignations from the order.
A member of the Order of Military Merit is given a certificate that bears his or her name, the signatures of the governor general and the secretary general of the order, and the seal of the order. To date there have been two versions of the certificate; the first was in a block font, while the second and present version is in script with the addition of "The Governor General of Canada, Chancellor of the Order of Military Merit" at the top of the document. The insignia was designed by Bruce Beatty.
Early designs for the Order of Military Merit.
Sovereign of the Order of Military Merit
Sovereign of the Order of Military Merit insignia.
Origins: As with the Order of Canada and most national honours, the Sovereign is presented with a special insignia.
Insignia: An 18-carat gold four-armed cross enamelled over scalloped machine work in translucent blue. Between each of the arms a rectangular diamond is set between two just-off-square diamonds. In the centre of the cross there is a maple leaf surmounted by an annulus, both of which are set with calibre rubies pavé. On the annulus the words MERIT — MERITE — CANADA appear in finely pierced gold. Above the cross there is a gold Royal Crown with seven diamonds set in the ermine, three in each of the fleurs-de-lys, and a larger one in the orb. On the arches are twenty-one pearls. The base is set with a sapphire, two emeralds, and two rubies. The cap is enamelled red.
Other: The Sovereign's insignia was manufactured by Garrard & Co. It is housed in a blue leatherette case lined with white satin and blue velvet. The Sovereign's badge was presented to the Queen in July 1973 by Governor General Roland Michener aboard the Royal yacht Britannia while it was anchored off Kingston, Ontario.
Chancellor of the Order of Military Merit
Origins: As with the Order of Canada and most national honours, the senior official is presented with insignia of office. Although a Chancellor's Chain and Principal Commander's Chain were proposed in 1972 when the Order of Military Merit was established, and designs were completed in 1975, it would not be until 1981 that the two chains were produced. This was largely due to the reluctance of then Governor General Jules Léger to proceed with the project, and it was only through the persistence of then Chief of the Defence Staff Admiral Falls that the project was finally executed.
Chancellor of the Order of Military Merit insignia.
Insignia: A chain of ten maple leaves alternating with four naval crowns (representing the navy), four mural crowns defaced with a maple leaf (representing the army), and four astral crowns composed of wings (representing the air force). The chain is completed by a centre device in the form of the shield from the Arms of Canada ensigned by the Royal Crown, each in their proper colours. A Commander's insignia is hung from this device. The entire chain was made of 9-carat gold and bears hallmarks for London, 1981. The Chancellor's Chain was made by Spink & Son of London. It is housed in a large blue leatherette case whose lid is embossed with the insignia of the order. The interior of the case is lined with white satin and red velvet.
Other: The Commander's insignia of the Order of Military Merit is detachable from the chain. The Chancellor's Chain is returned to the government upon retirement. The chain, like the other insignia of the order, was designed by Bruce Beatty.
Number: 9 (Edward Schreyer was the first to wear the Chancellor's Chain).
* • Right Honourable Roland Michener, PC, CC, CMM, OOnt, CD, QC: 1967–74.
* • Right Honourable Jules Léger, PC, CC, CMM, CD: 1974–79.
* • Right Honourable Edward Schreyer, PC, CC, CMM, OM, CD: 1979–84.
* • Right Honourable Jeanne Sauvé, PC, CC, CMM, CD: 1984–90.
* • Right Honourable Ray Hnatyshyn, PC, CC, CMM, CD, QC: 1990–95.
* • Right Honourable Roméo LeBlanc, PC, CC, CMM, ONB, CD: 1995–99.
* • Right Honourable Adrienne Clarkson, PC, CC, CMM, COM, CD: 1999–2005.
* • Right Honourable Michaëlle Jean, PC, CC, CMM, COM, CD: 2005–10.
* • Right Honourable David Johnston, CC, CMM, COM, CD: 2010 to present.
Principal Commander of the Order of Military Merit
Principal Commander of the Order of Military Merit insignia.
Origins: In most national military orders, the chief of the armed forces is usually an ex officio member. In Canada this means that the chief of the defence staff serves as the Principal Commander of the Order of Military Merit. Most chiefs of the defence staff have been either Officers or Commanders of the Order of Military Merit prior to their appointment as CDS. Nevertheless, they are presented with a special insignia for wear during investitures and state occasions.
Insignia: A chain of ten maple leaves alternating with four naval crowns (representing the navy), four mural crowns defaced with a maple leaf (representing the army), and four aerial crowns composed of wings (representing the air force). The chain is completed by a centre device in the form of the badge of the Canadian Armed Forces in its proper colours. The Principal Commander's insignia is hung from this device. The entire chain was made of 9-carat gold and bears hallmarks for London, 1981. The Principal Commander's Chain of the Order of Military Merit was made by Spink &Son. It is housed in a large blue leatherette case whose lid is embossed with the insignia of the order. The interior of the case is lined with white satin and red velvet.
Other: The Commander's insignia of the Order of Military Merit is detachable from the chain. The chain is returned to the government upon retirement. The chain, like the other insignia of the order, was designed by Bruce Beatty.
Numbers: There have been fifteen Principal Commanders and one acting Principal Commander, though the first Principal Commander to wear the chain was General R.M. Withers.
* • General Frederick Ralph Sharp, CMM, DFC, CD: 1972.
* • General Jacques Alfred Dextraze, CC, CBE, CMM, DSO, CD: 1972–77.
* • Admiral Robert Hilborn Falls, CMM, CD: 1977–80.
* • General Ramsey Muir Withers, CMM, CD: 1980–83.
* • General Gérard Charles Édouard Thériault, CMM, CD: 1983–86.
* • General Paul David Manson, OC, CMM, CD: 1986–89.
* • General Alfred John Gardyne Drummond de Chastelain, CC, CMM, CD, CH: 1989–93 and 1994–95.
* • Admiral John Rogers Anderson, CMM, CD: 1993–94.
* • General Joseph Édouard Jean Boyle, CMM, CD: 1995.
* • Vice-Admiral Lawrence Edward Murray, CM, CMM, CD (Acting): 1996–97.
* • General Joseph Gérard Maurice Baril, OC, CMM, CD: 1997–2001.
* • General Raymond Roland Joseph Henault, CMM, MSC, CD: 2001–05.
* • General Rickey John Hillier, OC, CMM, CD: 2005–08.
* • General Walter Natynczyk, CMM, MSC, CD: 2008–12.
* • General Thomas Lawson, CMM, CD: 2012–15.
* • General Jonathan Vance, CMM, MSC, CD: 2015 to present.
Commander of the Order of Military Merit
Commander of the Order of Military Merit.
Commander lapel pin.
Criteria: Appointments are made for outstanding meri-torious service in duties of great responsibility. This level is for those holding the rank of brigadier-general/commodore or above.
Insignia: A blue enamelled cross pattée in gold. In the centre of the insignia is a circlet surmounted by a Royal Crown with a red maple leaf in the centre. The circlet is inscribed with the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small ball in the centre through which a ring passes through which a 25 mm hanger passes. The insignia is worn around the neck.
Ribbon: A blue ribbon 38 mm wide edged by 5 mm of gold. On the undress ribbon, a Commander wears a small gold cross enamelled in blue and bearing a red maple leaf in the centre. Prior to 1983, a Commander simply wore a red maple leaf in the centre of the ribbon. Since that date, it is also permissible for a promoted member to wear the ribbon device related to a previous appointment in addition to the current device.
Lapel Badge: A small gold cross enamelled in blue and bearing a red maple leaf in the centre.
Postnominals: CMM.
Order of Military Merit 1st type certificate.
Order of Military Merit current certificate type.
Other: In terms of promotions within the order, only General Maurice Baril has held all three levels as well as having served as Principal Commander of the Order.
Number of Appointments: 262.
Officer of the Order of Military Merit
Officer of the Order of Military Merit.
Officer lapel pin.
Criteria: Appointments are made for outstanding meritorious service in duties of responsibility. This level is primarily for officers between the ranks of major or lieutenant-commander and colonel or captain(N). There have been instances where a chief warrant officer has been promoted to OMM because of the significance of his or her responsibilities.
Insignia: A blue enamelled cross pattée in gold. In the centre of the insignia is a circlet surmounted by a Royal Crown with a gold maple leaf in centre. The circlet is inscribed with the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small loop through which a ring passes attaching it to a laurelled bar. The OMM is worn on the left breast.
Ribbon: A blue ribbon 38 mm wide edged by 5 mm of gold. On the undress ribbon, an Officer wears a small gold cross enamelled in blue and bearing a gold maple leaf in the centre. Until 1983, an Officer wore a gold maple leaf in the centre of the ribbon. Since that date, it is also permissible for a promoted member to wear the ribbon device related to a previous appointment in addition to the current device.
Lapel Badge: A small gold cross enamelled in blue and bearing a gold maple leaf in the centre.
Postnominals: OMM.
Other: An issue number is typically engraved on the edge or the reverse of the lower arm.
Number of Appointments: 1,232.
Member of the Order of Military Merit
Member of the Order of Military Merit.
Member lapel pin.
Criteria: Appointments are made for exceptional service and performance of duty. It is normally for persons from non-commissioned ranks and junior officers below the rank of major or lieutenant-commander.
Insignia: A blue enamelled cross pattée in silver. In the centre of the insignia is a circlet surmounted by a Royal Crown with a silver maple leaf in the middle. The circlet is inscribed with the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small loop through which a ring passes, attaching it to a laurelled bar. The insignia is worn on the left breast.
Ribbon: A blue ribbon 38 mm wide edged with 5 mm of gold. On the undress ribbon, a Member wears a small silver cross enamelled in blue and bearing a silver maple leaf in the centre. Prior to 1983, a Member simply wore a silver maple leaf in the centre of the ribbon.
Lapel Badge: A small silver cross enamelled in blue and bearing a silver maple leaf in the centre.
Postnominals: MMM.
Other: Several different colour enamels have been used on this insignia, from deep Prussian blue to a dark, almost purple colour. An issue number is typically engraved on the edge or the reverse of the lower arm.
Spink standard MMM blue.
Spink Prussian blue MMM.
Spink & Sons hallmarks.
Number of Appointments: 3,049.
Insignia Manufacture: The following section outlines the various firms that have produced the insignia of the Order of Military Merit along with details related to dates of manufacture and placement of makers' marks. All insignia produced until 2014 were finished with vitreous enamels, while all insignia produced by the Royal Canadian Mint have a fill like hard epoxy.
Order of Military Merit Insignia Manufacturers and Material
Manufacturer | Commander | Officer and Member
---|---|---
Spink & Son | Marks: S&SN British hallmarks
Location: Reverse of the hanger and reverse of the lower arm
Material: 18-carat gold
Dates: 1972–84 | Marks: S&SN and British hallmarks
Location: Reverse of the suspender and reverse of the lower arm
Material: .925 silver (gold-plated for OMM)
Dates: 1972–84
Rideau Ltée | Marks: RIDEAU and 18 carat or RIDEAU + STER
Location: Reverse of the lower arm
Material: 18-carat gold from 1984–96, thereafter .925 silver, gold-plated
Dates: 1984–98 | Marks: RIDEAU + R STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OMM)
Dates: 1984–98
Birks
(Pressed Metal Products) | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver, gold-plated
Dates: 1998–2003, 2006 | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OMM)
Dates: 1998–2003, 2006–08
Pressed Metal Products | Marks: PMP STER
Location: Reverse of the lower arm
Material: .925 silver, gold-plated
Dates: 2011 | Marks: PMP STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OMM)
Dates: 2011, 2013*
*MMM insignia only
Bond Boyd Ltd. | Marks: BOND BOYD STER
Location: Reverse of the lower arm
Material: .925 silver, gold-plated
Dates: 2007–12 | Marks: BOND BOYD STER
Location: Reverse of the lower arm
Material: .925 silver (gold-plated for OMM)
Dates: 2009–11
Royal Canadian Mint | Marks: RCM logo
Location: Reverse of the lower arm
Material: .999 silver, gold-plated
Dates: 2014 to present
Other: Crown is set with jewels | Marks: RCM logo
Location: Reverse of the lower arm
Material: .999 silver (gold-plated for OMM)
Dates: 2014 to present
Numbering: Insignia made by Spink & Son were numbered on the reverse edge of the lower arm. All subsequent issues have been numbered on the reverse of the centre or the reverse of the lower arm with a three- or four-digit number.
Rideau hallmarks.
Birks hallmarks.
Bond Boyd hallmarks.
# 14
The Order of Merit of the Police Forces
Since the Order of Merit of the Police Forces is a recent addition to the family of Canadian honours, there is little direct history to be recounted about it. It is notable that Canada is one of the few countries to possess a separate order for its police services. In most other nations, outstanding service in the police force is recognized through the award of a civil order such as the Order of the British Empire in Britain or the Ordre national du Mérite in France.
Prior to the centennial of the Royal Canadian Mounted Police in 1973, there were questions concerning whether members of the country's various police forces could be appointed to the Order of Military Merit. The idea was quickly scrapped, though it was not without logic. In May 1935, Major-General James MacBrien, the commissioner of the RCMP, was made a Knight Commander of the Order of the Bath (military division). Thus, there was a precedent for military awards being used to recognize the police.
After the establishment of the Police Exemplary Service Medal in 1982, the Canadian Association of Chiefs of Police, which had lobbied for more than a decade, seemed satisfied with the single award, though by mid-1990 there was hope that a new civil order, aside from the Order of Canada, would be created and opened up to the police and fire services with the expectation that it would eventually include the other protective services (corrections, Coast Guard, emergency medical services, and peace officers) as partners. Instead, an order was established solely for the police, in part because of the reluctance of the fire services to become involved and also because the government did not wish to be seen as creating a broad new civil order that could potentially affect the public perception of the Order of Canada.
The new order was approved by the Queen in October 2000 and was officially announced by the governor general on March 19, 2001. Strangely, the order shares its insignia and appointment scroll with the Order of Military Merit. Other than the ribbon, the other difference is the insignia worn in the undress ribbon and as the lapel badge, which bears a natural maple leaf in the centre rather than a symmetrical one.
As with the Order of Canada and Order of Military Merit, the Queen is Sovereign of the order and the governor general serves as Chancellor during his or her term. The commissioner of the RCMP is the Principal Commander of the order during his or her time in office.
Appointees are selected by the Advisory Committee of the Order of Merit of the Police Forces, which consists of the president of the Canadian Association of Chiefs of Police (who acts as chair), the deputy commissioner of the RCMP, one person who is the head of a provincial police force (Ontario or Quebec), three municipal or regional chiefs of police, the president of the Canadian Police Association, and the deputy secretary of the Chancellery.
Provision is made for the appointment of one honorary Commander, Officer, and Member per year. The constitution of the order also sets out a yearly maximum number of awards: one-tenth of 1 percent of the average number of persons who were employees of police forces during the particular year. Within this quota are further limits on the types of awards. Elevation within the order is permitted. Unlike the Order of Miltiary Merit, no motto has yet been assigned to this order.
Sovereign of the Order of Merit of the Police Forces
Origins: As with the Order of Canada and most national honours, the Sovereign is presented with a special insignia. While a special Sovereign's insignia was made for both the Order of Canada and the Order of Military Merit, such an insignia has yet to be made for the Order of Merit of the Police Forces.
Insignia: None to date.
Chancellor of the Order of Merit of the Police Forces
Chancellor of the Order of Merit of the Police Forces insignia.
Origins: While every order has a titular head, it has also been a tradition for a senior official to be known as either a Grand Master or a Chancellor. As the governor general is the Queen's representative in Canada, it was decided that he or she should also serve as the Chancellor of the Order of Merit of the Police Forces and perform most of the investiture and administrative duties relating to certain aspects of the order. Following the precedent set by the Order of Canada, it was decided that Chancellor should be used for the second-highest official of the order.
Insignia: A neck chain composed of fourteen gold natural maple leaves and thirteen silver shields, alternating and joined by gold chain links. Each of the silver shields is defaced with a provincial or territorial flower in gold. The chain is completed by a centre device in the form of the shield from the Arms of Canada ensigned by the Royal Crown, each in their proper colours. A Commander's insignia is hung from this device. The Chancellor's Chain of the Order of Merit of the Police Forces was made by Pressed Metal Products.
Other: The insignia of the Order of Merit of the Police Forces is detachable from the chain. The Chancellor's Chain is returned to the government upon retirement.
Number: 3.
* • Right Honourable Adrienne Clarkson, PC, CC, CMM, COM, CD: 2000–05.
* • Right Honourable Michaëlle Jean, PC, CC, CMM, COM, CD: 2005–10.
* • Right Honourable David Johnston, CC, CMM, COM, CD: 2010 to present.
Principal Commander of the Order of Merit of the Police Forces
Principal Commander of the Order of Merit of the Police Forces insignia.
Origins: The Order of Merit of the Police Forces is based almost entirely on the Order of Military Merit. Thus, following this pattern, the head of the national police force (the Royal Canadian Mounted Police) is ex officio the Principal Commander of the Order of Merit of the Police Forces.
Insignia: A neck chain composed of fourteen gold natural maple leaves and thirteen silver shields, alternating and joined by gold chain links. Each of the silver shields is defaced with a provincial or territorial flower in gold. The chain is completed by a centre device from which a Commander's insignia is hung. The insignia was made by Pressed Metal Products of British Columbia.
Other: The chain is worn on special occasions such as investitures and the opening of Parliament.
Number: 3.
* • Commissioner Giuliano Zaccardelli, COM: 2000–06.
* • Commissioner William J.S. Elliott, COM, QC: 2007–11.
* • Commissioner Bob Paulson, COM: 2011 to present.
Commander of the Order of Merit of the Police Forces
Commander of the Order of Merit of the Police Forces obverse.
Criteria: The Commander level of the order is for outstanding meritorious service and demonstrated leadership in duties of great responsibility and over an extended period.
Insignia: A blue enamelled cross pattée in gold. In the centre of the insignia is a circlet surmounted by a Royal Crown with a red maple leaf in the centre. The circlet is inscribed with the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small ball in the centre through which a ring passes through which a 25 mm hanger passes. The insignia is worn around the neck.
Commander lapel pin.
Ribbon: Consists of equal proportions of blue, gold, and blue. On the undress ribbon, a Commander wears a small gold cross enamelled in blue and bearing a natural red maple leaf in the centre to differentiate it from the stylized maple leaf used by the Order of Military Merit.
Lapel Badge: A small gold cross enamelled in blue and bearing a natural red maple leaf in the centre.
Postnominals: COM.
Other: Of the maximum number of appointments that may be made to the order each year (one-tenth of 1 percent of those on police forces), no more than 6 percent may be made at the Commander level.
Number of Appointments: 22.
Officer of the Order of Merit of the Police Forces
Officer of the Order of Merit of the Police Forces obverse.
Officer lapel pin.
Criteria: Officers are appointed for outstanding meri-torious service in duties of responsibility over an extended period.
Insignia: A blue enamelled cross pattée in gold. In the centre of the insignia is a circlet surmounted by a Royal Crown with a gold maple leaf in the centre circumscribed by the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small loop through which a ring passes, attaching it to a laurelled bar. The insignia is worn on the left breast.
Ribbon: Consists of equal proportions of blue, gold, and blue. On the undress ribbon, an Officer wears a small gold cross enamelled in blue and bearing a natural gold maple leaf in the centre to differentiate it from the stylized maple leaf used by the Order of Military Merit.
Lapel Badge: A small gold cross enamelled in blue and bearing a natural gold maple leaf in the centre.
Postnominals: OOM.
Other: Of the maximum number of appointments that may be made to the order each year (one-tenth of 1 percent of those on police forces), no more than 30 percent may be made at the Officer level.
Number of Appointments: 161.
Member of the Order of Merit of the Police Forces
Member of the Order of Merit of the Police Forces obverse.
Criteria: Members are appointed for exceptional service or performance of duty over an extended period.
Insignia: A blue enamelled cross pattée in silver. In the centre of the insignia is a circlet surmounted by a Royal Crown with a silver maple leaf in the centre circumscribed by the words MERIT — MERITE — CANADA. The reverse bears a manufacturer's mark and an issue number.
Suspension: The top of the insignia has a small loop through which a ring passes, attaching it to a laurelled bar. The insignia is worn on the left breast.
Ribbon: Consists of equal proportions of blue, gold, and blue. On the undress ribbon, A Member wears a small silver cross enamelled in blue and bearing a natural silver maple leaf in the centre to differentiate it from the stylized maple leaf used by the Order of Military Merit.
Member of the Order of the Merit of the Police Forces reverse.
Member lapel pin.
Lapel Badge: A small silver cross enamelled in blue and bearing a natural silver maple leaf in the centre.
Postnominals: MOM.
Other: The annual limit is set at the maximum number of awards available to be conferred (one-tenth of 1 percent of those employed by police forces), less the number of Commanders and Officers appointed that year.
Number of Appointments: 519.
Insignia Manufacture: The insignia is struck in sterling silver and has been produced by a number of manufacturers, though insignia struck by Pressed Metal Products for Birks constitute more than three-quarters of the insignia awarded to date. All insignia produced until 2014 were finished with vitreous enamels, while all insignia produced by the Royal Canadian Mint have a fill like hard epoxy.
Order of Merit of the Police Forces Insignia Manufacturers and Material
Manufacturer | Commander | Officer | Member
---|---|---|---
Birks (Pressed Metal Products) | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver, gold-plated
Dates: 2001–06 | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver, gold-plated
Dates: 2001–08 | Marks: BIRKS STER
Location: Reverse of the lower arm
Material: .925 silver
Dates: 2001–08
Pressed Metal Products | None | None | Marks: PMP STER
Location: reverse of the lower arm
Material: .925 silver
Date: 2013
Bond Boyd Ltd. | None | None | Marks: BOND BOYD STER
Location: Reverse of the lower arm
Material: .925 silver
Dates: 2009–12
Royal Canadian Mint | Marks: RCM logo
Location: Reverse of the lower arm
Material: .999 silver, gold-plated for COM and OOM
Dates: 2014 to present
Other: Crown is set with jewels
Numbering: Each is numbered in the centre of the reverse with a two- or three-digit number prefixed with the letter P. The inclusion of the P was to prevent confusion with the insignia of the Order of Military Merit.
# 15
The Queen's Personal Honours
While all honours in Canada emanate from the Sovereign, the Queen has little direct involvement in who receives most Canadian honours. Although appointments to the Order of Canada and bestowal and all national honours are made in the name of the Queen, there is no known instance of the Queen personally nominating a person or requesting that a certain award be made or withheld in the Canadian context. This falls in line with the constitutional position of the Sovereign to advise her ministers but not to make actual policy.
Nevertheless, the Queen does possess four separate honours that she can directly bestow upon Canadians. These are the Order of Merit, the Royal Victorian Chain, the Royal Victorian Order, and the Royal Victorian Medal. Bestowal of these honours are made entirely on the Queen's own accord, and while suggestions for appointment to the Royal Victorian Order and the award of the Royal Victorian Medal occasionally emanate from the Canadian secretary to the Queen, the secretary to the governor general, or the Queen's private secretary, there is no formal nomination process. This unusual protocol is in place in part because of the highly limited number of awards bestowed and because of the type of service — personal service to the Queen or the Canadian Crown — being recognized. Consent from the Canadian government is usually sought when the Queen wishes to bestow the Order of Merit or the Royal Victorian Chain upon a Canadian.
The power to grant honours is part of the Royal Prerogative and the residual powers of the Crown. Over the past five hundred years the power of the Sovereign has gradually been eroded by elected officials, which also applies to the Sovereign's power to bestow honours. By the end of the nineteenth century, the Sovereign had lost control over almost all of the various orders of chivalry. Queen Victoria was unable to draw up honours lists, since this task had, through constitutional convention, been delegated to the prime minister and other senior ministers in the British government.
In 1896, Queen Victoria established the Royal Victorian Order as an order of chivalry over which she possessed complete control. This in part explains why it is so carefully and sparingly bestowed. The order consists of five levels: Knight/Dame Grand Cross, Knight/Dame Commander, Commander, Lieutenant (known until 1984 as Member Fourth Class or MVO IV), and Member (known until 1984 as Member Fifth Class or MVO V). Associated with this order is the Royal Victorian Medal, which is awarded in three grades: gold, silver, and bronze.
King Edward VII expanded on his mother's institution and established the Royal Victorian Chain and the Order of Merit in 1902. The Royal Victorian Chain is awarded as "a pre-eminent mark of the Sovereign's esteem and affection," while the Order of Merit is bestowed for "exceptionally meritorious service." These two awards remain among the most senior and most exclusive in the Commonwealth. Both are sparingly bestowed.
In 1946, King George VI also regained control over appointments to the Order of the Garter and the Order of the Thistle, the senior orders for England and Scotland respectively, generally bestowed on members of the nobility, though this practice has gradually changed over the past decades.
The Order of Merit and the Royal Victorian Chain have proven very useful in the Canadian context where titular honours such as knighthoods and peerages have been so controversial. On no fewer than two occasions, William Lyon Mackenzie King was offered a knighthood, but in each instance he refused. Mackenzie King had a curious phobia of honours. King George VI was well aware of this, yet he desperately wanted to recognize his Canadian prime minister after more than twenty years of exemplary service. Other leaders during the Second World War, such as Winston Churchill and South Africa's Jan Smuts, had been awarded the Order of Merit, and George VI decided that Mackenzie King should join them. After several days of contemplation and introspection, Mackenzie King decided to accept the Order of Merit. Churchill and Smuts would later joke with him that OM stood for "the Old Man Order."
The Royal Victorian Chain proved useful in the case of Vincent Massey, whom the Queen wished to appoint a Knight of the Garter, a move that would have made him the first from an overseas dominion. Both the Louis St. Laurent and John Diefenbaker governments indicated their disapproval of the proposal that the Queen make the award, which essentially put an end to the idea. In place of the Order of the Garter, then, Massey became only the second person of non-royal blood to be awarded the Royal Victorian Chain. It was Massey's most treasured honour.
The Royal Victorian Order was awarded to Canadians from its inception until 1946 when awards of it ceased. It was only in 1972 with the restructuring of the Canadian honours system that appointments to the Royal Victorian Order resumed. Appropriately, the first Canadian to receive the Royal Victorian Order after the restructuring was Esmond Butler, the secretary to the governor general, who had been instrumental in the creation of the modern Canadian honours system.
Canadians have also been appointed Knights Grand Cross and Knight Commanders of the Royal Victorian Order, though not for nearly a century. Lord Mount Stephen and Lord Strathcona, two of Canada's great railway builders, were both made GCVOs, while Lord Shaughnessy, another railway man, was made a KCVO. King Edward VII seemed to have had a real affection for Canadian railway barons! Canadians who went on to serve in the British government, such as Sir Edward Peacock and Sir Edwin Leather, were also admitted to the order as a GCVO and a KCVO respectively. Sir Conrad Swan, the Canadian-born former Garter King of Arms, was made a KCVO in 1994. During the 1939 Royal Visit of King George VI and Queen Elizabeth, Canada's governor general, Lord Tweedsmuir, was made a GCVO on the Royal Train as it travelled across Canada, making him the first person to have been knighted in a moving train.
Today the Order of Merit, the Royal Victorian Chain, the Royal Victorian Order, and the Royal Victorian Medal continue to be part of the Canadian honours system. Approximately, one Canadian is appointed to the Royal Victorian Order annually. During Royal Tours made by the Queen, there is often a small flurry of appointments. The 2010 Royal Tour resulted in twelve appointments to the Royal Victorian Order: two CVOs, one LVO, and nine MVOs.
The Order of Merit
Order of Merit obverse with Tudor Crown.
Origins: Founded by King Edward VII on September 23, 1902, the Order of Merit is the most exclusive non-titular order in the Commonwealth, being limited to twenty-four members. It has typically been reserved for those who have rendered "exceptionally meritorious service," and in the Canadian context this has usually meant most outstanding service of international magnitude.
Criteria: Any citizen of the Commonwealth qualifies who has rendered particularly outstanding service in the military or exceptionally meritorious service toward the advancement of art, literature, and science. In the Canadian context it has been used to recognize outstanding service that in other parts of the Commonwealth would usually be recognized through the bestowal of a knighthood.
Insignia: A gold cross of red and blue enamel. Within a wreath set upon a centre of blue enamel, the motto of the order, FOR MERIT, appears in gold letters. The reverse also bears a laurel wreath with the Royal Cypher in the centre. The entire insignia is surmounted by a gold Tudor Crown set with seed pearls. The insignia was originally manufactured by John Pinches of London. Military recipients of the order have a slightly different insignia in which crossed swords are added between the arms of the cross. To date there have been no awards of the military OM to a Canadian, though a number of governors general received the OM with swords, notably His Royal Highness the Duke of Connaught and Lord Alexander. The insignia presented to Jean Chrétien differs from the earlier issues in that it has a Royal Crown in place of the Tudor Crown.
Suspension: A ring passes through the top of the crown through which a 20 mm decorative suspension loop passes, through which the ribbon passes.
Ribbon: A ribbon 40 mm wide made up of Garter blue and Bath red.
Lapel Badge: None.
Postnominals: OM.
Other: Insignia issued after 1992 are returnable upon the death of the holder and each is engraved with a small number on the hinge of the crown. A recipient is also given a certificate, signed by the Queen, bearing his or her name and embossed with the seal of the order. This insignia is not worn in miniature.
Number of Canadian Appointments: 4.
Order of Merit, Queen Elizabeth II issue reverse.
* • Right Honourable William Lyon Mackenzie King, PC, OM, CMG: 1947.
* • Dr. Wilder Penfield, OM, CC, CMG, CD: 1953.
* • Right Honourable Lester B. Pearson, PC, OM, CC, OBE: 1972.
* • Right Honourable Jean Chrétien, PC, OM, CC, QC: 2009.
The Royal Victorian Chain
Origins: Like the Order of Merit, the Royal Victorian Chain was established by King Edward VII in 1902.
Criteria: Awarded by the Queen as a pre-eminent mark of the Sovereign's esteem and affection toward such persons as Her Majesty especially desires to honour.
Insignia: The chain consists of roses, trefoils, thistles, and lotuses (two of each and one extra rose) connected by a double gold chain consisting of seven links. In the centre of the chain is the Royal Cypher of King Edward VII (ERI) enamelled in red and surmounted by the Imperial Crown. From this hangs a white Maltese cross of eight points, the centre of which contains an oval surrounded by a blue border containing the word VICTORIA. The centre of the oval is crimson enamel, in the centre of which is the Royal Cypher of Queen Victoria (VRI). Both the crown and Queen Victoria's Royal Cypher are set in diamonds.
Royal Victorian Chain obverse.
Royal Victorian Chain with numbering on reverse.
Suspension: A small eyelet connects the insignia of the Maltese cross to the actual chain.
Ribbon: For men there is no ribbon associated with this honour. For women the ribbon of the Royal Victorian Order is fashioned into a bow, along with four elements of the chain, which is worn on the left.
Lapel Badge: None.
Postnominals: None.
Other: More than ninety-five of the awards made have been to foreign monarchs or foreign heads of state. Vincent Massey's award was only the second made to a commoner. Roland Michener's was the third; his chain was number fifty-two. A recipient of the chain is also given a warrant, signed by the Queen, bearing his or her name. This insignia is not worn in miniature.
Number of Conferrals to Canadians: 2.
* • Right Honourable Vincent Massey, PC, CH, CC, CD: 1960.
* • Right Honourable Roland Michener, PC, CC, CMM, OOnt, CD, QC: 1974.
The Royal Victorian Order
Origins: Founded by Queen Victoria in 1896, the Royal Victorian Order consists of five levels: Knight/Dame Grand Cross, Knight/Dame Commander, Commander, Lieutenant, and Member. At present Canadians are ineligible for the two senior-most levels of this order because they confer knighthood. Appointments listed are for service rendendered in Right of Canada. A number of Canadians — notably Sir Edwin Leather, KCMG, KCVO, and Sir Conrad Swan, KCVO, KGCN — were appointed to the titular level of the order for service in Right of the United Kingdom.
Criteria: There are no set criteria for the Royal Victorian Order other than extraordinary or personal services to the Sovereign or to the Royal Family. CVOs are typically reserved for senior officials such as the secretary to the governor general and organizers of Royal Tours. LVOs have been granted to long-serving private secretaries to lieutenant governors and long-serving aides-de-camp. MVOs are usually for protocol officers and those serving as the Queen's Equerry.
All appointments to the order are accompanied by a warrant of appointment bearing the name of the recipient and the signatures of the Queen and the grand master of the order and are embossed with the seal of the order.
Insignia: Varies depending on the level of appointment. For women appointed GCVO and DCVO, their insignia is of a slightly reduced size.
Lapel Badge: A small rosette in the colours of the ribbon of the order with the Royal Cypher of Queen Victoria (VRI) in gold in the centre. The same lapel badge is worn by all members of the order.
Knight/Dame Grand Cross of the Royal Victorian Order
Knight/Dame Grand Cross of the Royal Victorian Order.
Mantle: Dark blue satin edged in red satin 6 cm in width, the edging of the collar being white silk 1.5cm in width, the mantle is fastened with a cordon of dark blue silk and gold, with two tassels of the same colour. Embroidered in wire on the left side is a representation of the breast star of a GCVO.
Collar: Gold in colour composed of octagonal pieces displaying a gold rose, alternating with pierced ornamental frames, each containing a portion of Queen Victoria's Latin style VICTORIA — BRITT — DEF. FID — IND. IMP. The central device of the collar from which the sash badge may be hung consists of an octagon enamelled blue, edged with red, and charged with a white saltire, in the centre of which is a gold medallion bearing a crowned effigy of Queen Victoria.
Sash Badge: An eight-point white-enamelled Maltese cross with a gold frame 95 mm in width in the centre of which is an oval surrounded by a blue border containing the word VICTORIA. The centre of the oval is crimson enamel, in the centre of which is the Royal Cypher of Queen Victoria (VRI). The reverse is plain and stamped with an inventory number.
Breast Star: A silver eight-point chipped star, in the centre of which is a reduced version of the GCVO sash badge in gold with enamels.
Ribbon: Blue in colour, edged on each side by stripes of red, white, and red, and worn over the right shoulder, 9 cm in width for men and 6.75cm in width for women.
Postnominals: GCVO.
Number of Appointments to Canadians: 2.
* • Lord Mount Stephen, Bt, GCVO, Kt: 1905.
* • Lord Strathcona, PC, GCMG, GCVO: 1908.
Knight/Dame Commander of the Royal Victorian Order
Knight/Dame Commander of the Royal Victorian Order.
Neck Badge: An eight-point enamelled white Maltese cross with a gold frame 50 mm in width, in the centre of which is an oval surrounded by a blue border containing the word VICTORIA. The centre of the oval is crimson enamel, in the centre of which is the Royal Cypher of Queen Victoria (VRI). The reverse is plain and stamped with an inventory number.
Breast Star: A silver chipped Maltese cross with small rays issuing from the centre, in the centre of which is a reduced version of the neck badge with arms in frosted silver, the remainder enamelled (the central device is the same as an MVO insignia).
Suspension: A small eyelet is mounted at the apex of the uppermost arm of the Maltese cross through which passes a 20 mm hanger and ribbon.
Ribbon: Blue in colour, 38 mm wide, and edged on each side with 2 mm of red, 2 mm of white, and 2 mm of red.
Postnominals: KCVO or DCVO.
Number of Appointments to Canadians: 1.
* • Lord Shaughnessy, KCVO, Kt: 1908.
Commander of the Royal Victorian Order
Commander of the Royal Victorian Order obverse.
Insignia: The same as the neck insignia worn by a Knight Commander, but no breast star is worn by Commanders.
Suspension: The same as for a Knight Commander.
Ribbon: The same as for a Knight Commander.
Postnominals: CVO.
Number of Appointments: 17 (pre-1946), 54 (post-1972).
Lieutenant of the Royal Victorian Order
Lieutenant of the Royal Victorian Order obverse.
Insignia: An eight-point Maltese cross with a gold frame 44.5 mm in width and filled with white enamel, in the centre of which is an oval surrounded by a blue border containing the word VICTORIA. The centre of the oval is crimson enamel, in the centre of which is the Royal Cypher of Queen Victoria (VRI). The reverse is plain and stamped with an inventory number.
Suspension: A small eyelet is mounted at the apex of the uppermost arm of the Maltese cross through which passes a ring and ribbon. The LVO is worn on the left breast.
Ribbon: Blue in colour, 32 mm wide, and edged on each side with 1 mm of red, 1 mm of white, and 1 mm of red.
Postnominals: LVO.
Number of Appointments: 14 (pre-1946), 83 (post-1972).
Member of the Royal Victorian Order
Member of the Royal Victorian Order obverse.
Member of the Royal Victorian Order reverse.
Insignia: An eight-pointed Maltese cross, 41 mm in width, in frosted silver, the centre of which contains a gold oval surrounded by a blue border containing the word VICTORIA. The centre of the oval is crimson enamel, and in its centre is the gold Royal Cypher of Queen Victoria (VRI). The reverse is plain and stamped with an inventory number.
Royal Victorian Order certificate.
Suspension: A small eyelet is mounted at the apex of the uppermost arm of the Maltese cross through which passes a ring and ribbon. The MVO is worn on the left breast.
Ribbon: Blue in colour, 32 mm wide, and edged on each side with 1 mm of red, 1 mm of white, and 1 mm of red.
Postnominals: MVO.
Number of Appointments: 1 (pre-1946 to a Newfoundlander), 75 (post-1972).
Insignia Manufacture: The insignia of the order has been manufactured by a number of firms.
Manufacturer | Dates | Classes
---|---|---
Collingwood & Co Ltd. | 1896–2001 | All classes
Toye, Kenning & Spencer | 2002–04
2012 to present | All classes
All classes
Cleave | 2005–08
2009–12 | All classes
LVO and MVO
Thomas Fattorini | 2008 to present | KCVO, DCVO, CVO
Worcester Medal Service | 2012 to present | LVO and MVO
Numbering: It was not until after the death of Queen Victoria in 1901 that the insignia began to be numbered.[1] The numbers have been used primarily for inventory purpose, and there is no indication that there was ever a desire to track the insignia and to whom it had been awarded. There are a variety of prefixes and suffixes that have been stamped into the centre of the reverse of all sash badges, neck badges, breast stars, and breast insignia manufactured after 1901. The numbers on the reverse of GCVO, KCVO, and DCVO sets always match. Thus, a GCVO sash badge and breast star set bear the same number — for example, 17. For KCVOs the prefix K and inventory number are stamped in the centre of the reverse of the neck badge and breast star — for example, K225. For CVOs the neck badge is numbered in the centre of the reverse with the prefix C and the inventory number — for example, C335.[2] The numbering of the MVO (IV) or LVO and MVO (V) or MVO insignia bears a number impressed in the centre of the reverse. There is at least one known case of an insignia that is stamped with DUPLICATE below the inventory number.[3]
When the statutes of the order were changed in 1936 to allow for the appointment of women to the order, another numbering differentiation was introduced. GCVO insignia bestowed on women were numbered the same as that for men. For Dame Commanders the numbering was preceded with a D and an inventory number — for example, D123. For Commanders the inventory number was followed by the letter L for "Lady" — for example, C14L. For Member (IV) or LVO and MVO (V) or MVO, the inventory number was preceded by the letter L — for example, L77.
The Royal Victorian Medal (Gold, Silver, and Bronze)
Royal Victorian Medal in silver obverse.
Royal Victorian Medal in silver reverse.
Origins: Established by Queen Victoria in 1896 as a reward for servants of the Royal Household who had rendered personal service to the Queen and the Royal Family, the Royal Victorian Medal has three levels: gold, silver, and bronze. The medal is part of the Royal Victorian Order, though until 1984 recipients were not entitled to use any postnominals.
Criteria: Available for servants of the Royal Household (including Government Houses), junior civil servants such as drivers and housekeepers, and non-commissioned members of the Canadian Armed Forces, who have rendered personal service to the Queen and the Royal Family. It is awarded at the discretion of the Sovereign.
Insignia: A 28 mm circular medal struck in gold (silver gilt), silver, and bronze, depending on the level of award. The obverse bears an effigy of the Queen surrounded by ELIZABETH II DEI GRATIA REGINA F:D:. The reverse of the medal bears an uncrowned Royal Cypher surrounded by ribbons and laurels, with ROYAL VICTORIAN MEDAL upon a curled ribbon in the base. The medals are struck at the Royal Mint in London.
Royal Victorian Medal certificate.
Suspension: A small eyelet is screwed into the top of the disc through which passes a ring.
Ribbon: 32 mm wide blue ribbon edged on each side with 2 mm of red, 2 mm of white, and 2 mm of red.
Bars: Additional awards can be made, and bars have been awarded to the silver and bronze medals, though no bars have yet been issued for the gold.
Royal Victorian Order/Royal Victorian Medal lapel pin.
Naming: None.
Lapel Badge: Same as that worn by members of the order.
Postnominals: RVM.
Other: A recipient can be awarded the gold, silver, and bronze RVM and wear all of them simultaneously. He or she may also be appointed to the order and continue to wear the RVM(s) along with the insignia of membership.
Number Awarded: 2 (pre-1946), 22 (post-1972).
# 16
The Most Venerable Order of the Hospital of St. John of Jerusalem
Along with the Royal Victorian Order, the Most Venerable Order of the Hospital of St. John of Jerusalem (the Order of St. John) constitutes one oof the oldest components of the Canadian honours system. It predates even the Royal Victorian Order, but it was not until 1990 that the Order of St. John was formally incorporated into the Canadian honours system.
The order has played an important role in the early development of first-aid training in Canada and also the Canadian honours system. Indeed, it was inquiries from officials of the order that helped prompt Prime Minister R.B. Bennett to reinstitute the honours list in the early 1930s.
The Order of St. John "is a working order of charitably minded men and women whose philanthropy is expressed principally through its two foundations, the St. John Eye Hospital in Jerusalem and the St. John Ambulance Association and Brigade."
The original Order of St. John of Jerusalem had been established in England during the twelfth century after the First Crusade to maintain a hospital in Jerusalem, but King Henry VIII passed a law dissolving it in 1540. A number of attempts were made to revive the order, primarily in France.
The earliest origins of the order in Canada can be traced back to New France and Acadia. Knights of Malta (i.e., of the Sovereign and Military Order of Malta) held a variety of high offices in the colony, including at various times the governor of Quebec and the lieutenant governor of Acadia. The order ceased to operate in Canada after the fall of New France in 1760.
In 1831 the Grand Priory of the British Realm of the Most Venerable Order of the Hospital of St. John of Jerusalem was re-established, while the St. John Ambulance Association, with the purpose of offering first-aid training and transportation of the sick and wounded, came into existence in 1877. With this development the order found a new purpose and membership grew steadily. The association also gave rise to the St. John Ambulance Brigade. Queen Victoria granted the order a Royal Charter in 1888, and with this the Sovereign became the Sovereign Head of the Order and Patron.
The Canadian branch of the order got an informal start in 1883 when an inaugural first-aid course was held in Quebec City. In March 1895, Canada's deputy surgeon general, George Sterling Ryerson, was authorized by the Grand Priory to establish a branch of the St. John Ambulance Association in Canada. Sir George Airy Kirkpatrick, lieutenant governor of Ontario, became the association's first presi-dent and played an important role in establishing the order in Canada. The order's first Canadian ambulance division was established in London, Ontario, in 1909.
When the Nickle Resolution was adopted in 1918, all honours came to a virtual end in Canada — and as a September 15, 1919, memorandum written by the minister of justice confirms, this included awards of the Order of St. John. The award of the order had served as a reward for devoted volunteers; thus, the organization was denied an important aspect of its operation. During the premiership of William Lyon Mackenzie King, a number of unofficial inquiries were made concerning the status of the order and whether or not it could still be awarded to Canadians. Mackenzie King, reluctant to deal with the honours issue, merely cited the Nickle Resolution as a blanket prohibition on all honours, including the Order of St. John.
In October 1930, after his election as prime minister, R.B. Bennett received a letter from the Sub-Prior of the Order, the Earl of Scarborough, who explained the nature of the order's work in Canada and why he thought the Nickle Resolution was never intended to prevent appointments to the order. By November of the same year, after discussing the matter with the order's Grand Prior, the Duke of Connaught (himself a former governor general of Canada), Bennett decided "it was absurd that you [the Order of St. John] have been treated as you have." In essence, this signalled Bennett's agreement to allow the resumption of appointments to the order in Canada. Despite this tacit assent, however, Bennett did not immediately change the government's policy toward the order, since he was planning to revive the annual honours list and not just the Order of St. John.
By June 21, 1931, only a day before the order was to celebrate the centenary of its revival in England, Bennett cabled Scarborough: "After careful consideration our Government has concluded that appointments in your ancient Order are not subject to the restrictions of the resolution of the House of Commons, and we will accept responsibility for any Canadian appointments." Canadian appointments to the order resumed after Bennett's intervention. The very next day Scarborough sent a message to Bennett, thanking him for looking into the matter and also informing the prime minister that a number of Canadians were to be admitted to the order as part of the centennial celebrations.
More than a year later, on October 20, 1932, Bennett received notice that the Grand Priory wished to appoint him a Knight of Grace. Bennett accepted the honour, one that recognized his role in helping the order in Canada as well as his position as prime minister of Canada. Bennett's involvement in the order was not fleeting; in fact, he went on to represent Canada at a number of conferences.
The resumption of Canadian appointments to the Order of St. John to Canadians resulted in the establishment of the Commandery of Canada in 1934. By 1946 the Canadian branch was elevated to a full Priory, with the governor general serving as the Prior of Canada.
The Order of St. John was reorganized in 1999, and the Priories in Canada, Australia, New Zealand, South Africa, Scotland, Wales, and the United States were put on an equal footing with the Priory in England. The Secretariat of the Order of St. John came into existence in October 1999.
At the international level the Grand Prior is appointed by the Sovereign Head of the Order after consultation with the Grand Prior's Advisory Council. The Grand Council consists of the various Great Officers: the Lord Prior, the Prelate, the Deputy Lord Prior, the Sub-Prior, and a senior representative from each of the eight Priories. Priories exist in Australia, Canada, New Zealand, South Africa, England, Scotland, Wales, and the United States. There are also thirty branches of the order in various Commonwealth countries, and in the Republic of Ireland the order is considered an associate body.
The Queen is the Sovereign Head of the Order, while the Duke of Gloucester serves as Grand Prior. The governor general serves as Prior of the Order in Canada, and at that time he or she is simultaneously appointed a Knight or Dame of Justice. Each lieutenant governor and territorial commissioner serves as Vice-Prior, and those who accept are appointed as Knights/Dames of Grace. Since 1990 the order has been fully constituted as part of the Canadian honours system.
All levels of the Order of St. John are available to any citizen who has performed good service for the order and who undertakes to conform to the rules of the order. Since 1992 there has been no religious requirement for membership. Mantles may be worn on special occasions such as investitures by all those holding appointments of Officer and above.
The order has various councils across Canada, and nominations for admission to or promotion within the order are initiated at the branch level before they pass on to the provincial council. From there they are forwarded to the national level where they are considered by the awards committee. The awards committee then transmits its list to the governor general, who approves the nominations on behalf of the Queen. The annual number of appointments to and promotions within the order ranges from seventy to ninety. The curtailing of large annual appointments began in the early 2000s as more rigorous nomination criteria were introduced by St. John Ambulance across Canada.
The Priory of Canada maintains fraternal relations with two other organizations that trace their origins to the original Hospitallers in the Holy Land: the Canadian Association of the Sovereign Military Order of Malta (generally referred to as the Order of Malta) and the Canadian Association of the Johanniter Orden. Although these orders are not part of the Canadian honours system, they are well established internationally, and they recognize each other as being part of a common patrimony. There are several other St. John orders active in Europe in this alliance, though they do not have an organized presence in Canada. Other organizations use the name St. John (or Knights of Malta, Knights Hospitaller, and so forth), but these are not recognized as legitimate by the St. John orders mentioned above.
Sovereign Head of the Order of St. John
Sovereign of the Order of St. John sash badge.
Sovereign of the Order of St. John breast star.
Origin: As with many of the other orders in the British and Canadian honours systems that are national/Commonwealth in scope, the Sovereign wears a distinctive insignia to denote her position as head and proverbial fountain of all honours.
Insignia: The breast star is an eight-pointed cross 8 cm wide set in gold with enamels. The sash badge is an eight-pointed cross 8 cm wide set in gold with embellishments surmounted by a St. Edward's Crown, which is not enamelled.
Other: For a queen the sash badge is hung from a black watered ribbon 6 cm wide. For a king the sash is 10 cm wide.
Number: 6.
* • Victoria: 1883–1901.
* • Edward VII: 1901–10.
* • George V: 1910–36.
* • Edward VIII: 1936.
* • George VI: 1936–52.
* • Elizabeth II: 1952 to present.
Prior of the Order of St. John
Prior of the Order of St. John insignia.
Origin: As with the Order of Canada, the Order of Military Merit, and the Order of Merit of the Police Forces, senior officials of a number of the national orders wear distinctive chains of office to denote their positions within their orders. The Prior of Canada of the Order of St. John thus decided to follow the pattern set by other Canadian orders and established special insignia for its senior officials.
Insignia: The insignia is composed of two separate rows of gold chain joined in the centre by the shield of Canada, which is surrounded by the motto of the Order of Canada and surmounted by a Royal Crown. From this hangs a gold eight-pointed cross in white enamel defaced with the shield of the Priory of Canada in red and white enamel. On the chain are the thirteen shields of the Canadian provinces and territories in gold with enamel, alternating with gold and white enamel eight-pointed crosses. Each side of the chain bears the shield of the Priory of Canada in gold with red and white enamels at the point where the chain affixes to the shoulder.
Other: The concept for the chains was developed by Roger Lindsay and Stephen Connelly of Cleave & Company.[1] Manufactured by Cleave & Company in 2002, the Prior's Chain was donated to the Priory of Canada by Roger Lindsay.
Number: Governor General Michaëlle Jean was the first Prior to be presented with the Prior's Chain. It had previously been worn as the Chancellor's Chain by Chancellor Jeff Gilmour, but in 2006 it was decided that this chain would more appropriately be worn by the governor general as Prior of the Order, and so the chain was changed into the Prior's Chain of Office and was formally presented to Jean by then Lord Prior Eric Barry at Rideau Hall on April 12, 2006. While there have been twelve Priors of the Order of St. John since the establishment of the Priory of Canada in 1946, only two have worn the Prior's Chain.
* • Right Honourable Michaëlle Jean, PC, CC, CMM, COM, DStJ, CD: 2006–10.
* • Right Honourable David Johnston, CC, CMM, COM, KStJ, CD: 2010 to present.
Chancellor of the Order of St. John
Early Order of St. John certificate.
Origins: The Chancellor's Chain was created at the same time as the Prior's Chain of office, though it was originally used by the Past Chancellor.
Insignia: A chain composed of two separate rows of gold chain joined together in the centre by the shield of Canada, which is surrounded by the motto of the Order of Canada and surmounted by a Royal Crown. From this hangs a gold eight-pointed cross in white enamel defaced with the shield of the Priory of Canada in red and white enamel. An immediate Past Chancellor's Chain of Office also exists, which is identical to the Chancellor's Chain with the exception that there are no gold maple leaves on each side of the shield of Canada.
Later issue Order of St. John certificate.
Current issue Order of St. John certificate.
Other: As with the Prior's Chain, the Chancellor's Chain was made by Cleave & Company and was donated by Roger Lindsay.
Number: While there have been twenty Chancellors of the Order of St. John since the Priory of Canada was established in 1946, only those since 2006 have worn the Chancellor's Chain.
* • The Honourable René Marin, CM, OMM, OOnt, KStJ, CD, QC: 2005–07.
* • John C. Mah, GCStJ, CD, QC: 2007–09.
* • Richard Bruce, MOM, GCStJ: 2009–11.
* • Mairi Arthur, DStJ: 2011–14.
* • Robert White, KStJ: 2014 to present.
Bailiff or Dame Grand Cross
Bailiff Grand Cross of the Order of St. John.
Criteria: Pre-eminent contribution to the order and St. John Ambulance at the national and international levels.
Mantle: Black silk lined with black silk and on the breast the badge of the order in gold wire and white material.
Sash Badge: For men there is a sash worn from the right shoulder to the left hip from which is hung a white enamel Maltese cross set in gold with alternating lions and unicorns between each arm. The insignia measures 85 mm across; for women it is slightly smaller, measuring 59 mm across.
Breast Star: The breast star measures 92 mm across and is a white enamel Maltese cross set in gold-coloured metal.
Suspender: The sash badges are usually attached by a small ring at the top of the Maltese cross, which is connected to another ring that is clipped onto the sash below the bow. The breast star is attached by a straight pin mounted on the back of the star.
Ribbon: The ribbon is black moiré. For the men's sash it is 120 mm in width; for women it is 58 mm wide.
Lapel Badge: A black rosette with a white-and-gold Maltese cross in the centre.
Postnominals: GCStJ (used within the order only).
Other: A variety of insignia exist ranging from high-quality versions set in 18-karat gold, or silver gilt to the more recent gold-plated base metal.
Number of Appointments: Ten Canadians have been appointed to the order's highest grade. Generally, there are only twelve male and seven female members of the order at this grade for the entire Commonwealth.
* • Right Honourable Vincent Massey, PC, CH, CC, GCStJ, CD: 1955.
* • Mrs. Margaret MacLaren, GCStJ: 1963.
* • Commissioner Leonard Nicholson, OC, MBE, GCStJ: 1971.
* • Mrs. Kathleen Gilmour, MBE, GCStJ: 1975.
* • Brigadier-General Cyrille J. Laurin, OBE, GCStJ: 1980.
* • Mlle. Yvette Loiselle, GCStJ: 1983.
* • Brigadier-General Edward Beament, OBE, CM, GCStJ, ED, CD, QC: 1995.
* • Lieutenant-Colonel Eric Barry, GCStJ, CD: 1999.
* • John Chew Mah, GCStJ, CD, QC: 2009.
* • Richard Bruce, MOM, GCStJ: 2011.
Knight or Dame of Grace or of Justice
Knight/Dame of Justice of the Order of St. John.
Criteria: Appointments are made to those who have successfully demonstrated leadership at a national or regional level in a position carrying major responsibility or have demonstrated exceptional leadership and performed meritorious service at a local level.
Mantle: Black merino faced with black silk bearing an embroidered representation of the breast star of the order at the appropriate grade. The same style mantle with varying size embroidered Maltese crosses may be worn by members of the order beginning with the rank of Officer.
Neck Insignia: For men a white enamel Maltese cross 58 mm across set in silver for Knights or Dames of Grace and gold for Knights or Dames of Justice with alternating lions and unicorns between each arm. Women wear their "neck" insignia from a bow.
Breast Star: A white enamel Maltese cross 78 mm across set in silver (with alternating lions and unicorns between each arm) for Knights or Dames of Grace and gold (without the lions and unicorns) for Knights or Dames of Justice.
Suspender: For the neck badge a small eyelet is mounted on top of the insignia through which passes a 25 mm long suspender bar connected to the neck ribbon. The breast star is attached by a straight pin mounted on the back of the star.
Ribbon: Black moiré 38 mm in width.
Lapel Badge: A black rosette with a white-and-gold Maltese cross in the centre.
Postnominals: KStJ or DStJ (used within the order only).
Knight/Dame of Grace of the Order of St. John (breast star).
Knight/Dame of Grace of the Order of St. John (neck badge).
Other: A person who is a Knight or Dame of Grace can be reclassified as a Knight or Dame of Justice if he or she can prove to the Genealogist of the Priory that he or she is entitled to bear arms or is a Prior of a Priory.
Number of Appointments: Approximately four to six appointments annually.
Commander
Commander of the Order of St. John obverse.
Order of St. John Commanders and Officers lapel pin.
Criteria: Appointments are made to those who have successfully demonstrated leadership at a regional level in a position carrying substantial responsibility or have demonstrated exceptional leadership and performed meritorious service at a local level.
Insignia: A 58 mm wide Maltese cross with a silver frame enamelled in white with alternating lions and unicorns between each arm.
Suspender: A small eyelet is mounted on top of the insignia through which passes a 25 mm long suspender bar connected to the neck ribbon.
Ribbon: Black moiré 38 mm in width.
Lapel Badge: A small silver Maltese cross with white enamel.
Postnominals: CStJ (used within the order only).
Number of Appointments: Approximately fifteen appointments annually.
Officer
Officer of the Order of St. John obverse.
Criteria: Appointments are made to those who have successfully demonstrated leadership at a regional level in a position carrying substantial responsibility or at a local level carrying major responsibility.
Insignia: A 45 mm wide Maltese cross with a silver frame enamelled in white with alternating lions and unicorns between each arm.
Suspender: A small eyelet on the insignia is connected to a larger silver ring through which passes the ribbon.
Ribbon: Black moiré 38 mm in width.
Lapel Badge: The same as that worn by Commanders.
Postnominals: OStJ (used within the order only).
Number of Appointments: Approximately twenty to twenty-five appointments annually.
Serving Member
Serving Member of the Order of St. John obverse.
Serving Brother/Sister of the Order of St. John frosted issue.
Criteria: Appointments are made to those who have performed good and devoted service to the order through sustained voluntary commitment of time, skills, and/or resources to St. John.
Insignia: There have been a wide variety of insignia used for this level of the order over the past century. Currently, the insignia measures 38 mm across and is a bright plain silver Maltese cross with alternating lions and unicorns between each arm.
Serving Brother/Sister of the Order of St. John circular issue obverse (multi-piece construction).
Serving Brother/Sister of the Order of St. John circular issue obverse (multi-piece construction with separate centre).
Suspender: A small eyelet on the insignia is connected to a larger silver ring through which passes the ribbon.
Ribbon: Black moiré 38 mm in width.
Lapel Badge: A small silver Maltese cross.
Postnominals: MStJ (used within the order only).
Other: This level was formerly classified as Serving Brother or Serving Sister of the Order of St. John.
Serving Brother/Sister of the Order of St. John wartime economy issue obverse.
Number of Appointments: Approximately forty to sixty appointments annually.
Serving Member lapel pin.
Insignia Manufacture: The insignia has been made by a number of firms: Lamb & Company, Toye, Kenning & Spencer, Spink & Son, Joe Drouin Ltd., Worcester Medal Service, and Cleave & Company. A variety of materials have been used to manufacture the insignia: sterling silver with enamels, silver-plated or rhodium-plated base metal with enamels, and base metal with epoxy fill.
# 17
The Provincial Orders
The Origins of the Provincial Orders
The establishment of provincial orders is not an entirely recent event. However, their acceptance and incorporation into the broader Canadian Order of Precedence is. Details about the history of the provincial orders are scarce, primarily because, as most were founded in the past two decades, documents related to their creation remain sealed.
The first province to found an order was Quebec: the Order of Agricultural Merit of the Province of Quebec in 1925. The origins of this order can be traced back to 1869 and the passage of the Department of Agriculture and Colonization Act, which charged the Department of Agriculture with the twin goals of ensuring the growth and success of agricultural endeavours and increasing immigration to Quebec.
The act also created a Council of Agriculture composed of twenty-three Quebec farmers and agricultural experts who were given the responsibility of organizing agricultural societies in each county and district and holding competitions. These competitions were designed "t]o encourage, by means of special prizes, the cultivation of certain variety of plants and grain, which it is desirable to extend throughout the province."[[1] Initially, the Quebec Agricultural Merit Awards were in the form of large bronze medals, 50 mm in diameter, depicting the arms of Quebec on the obverse and engraved with the name of the recipient on the reverse.
Early Small Quebec Agriculture Medal.
In 1889 the Agricultural Merit Awards were standardized and organized into four levels: Distinguished Exceptional Merit (gold medal), Distinguished Merit (silver), Great Merit (bronze), and Merit (diploma/certificate).
Quebec Agriculture Merit Medal in bronze.
The gold medal was awarded to the farmer who had "most distinguished himself" during the five years since receiving the silver medal. The silver medal was awarded to a person who had "obtained in the competition 85 out of 100 points allowed for perfect cultivation." The bronze medal was awarded for seventy-five points "for perfect cultivation," and the diploma was awarded for a score of sixty-five points.
The diplomas and certificates were regulated by the lieutenant governor-in-council (the provincial Cabinet) and were changed from time to time. Provisions were also made to allow for persons based outside of Canada to receive awards in recognition of "public service or in scientific missions or by works or publications on agriculture." Quebec's minister of agriculture was an ex officio holder of the gold medal.
This entire system was modernized on April 3, 1925, with the creation of the Order of Agricultural Merit of the Province of Quebec instituted for the "purpose of encouraging agriculturalists by honours and rewards, and to acknowledge services rendered to agriculture."
The new order was divided into three levels: the Commander of the Order of Agricultural Merit of Quebec (Distinguished Exception Merit, formerly the gold medal), Officer of the Order of Agricultural Merit of Quebec (Distinguished Merit, formerly the silver medal), Knight/Chevalier of the Order of Agricultural Merit (Great Merit, formerly the bronze medal), as well as the Diploma of Merit.
Quebec Commander of the Order of Agricultural Merit obverse.
The lieutenant governor-in-council was responsible for setting out the conditions under which competitions for these awards were held. Judges for competitions were appointed by the minister of agriculture and were selected from the membership of the order (existing Commanders, Officers, and Chevaliers). The provision to award the order for scholarship or research was retained. As well, holders of the Quebec Agricultural Merit Awards were transformed into holders of the Quebec Order of Agricultural Merit at the appropriate levels. It would appear that the minister of agriculture became an ex officio Commander of the order, as well.
Quebec Officer of the Order of Agricultural Merit obverse.
This order was based on France's Ordre du Mérite agricole and highlights the importance of agriculture to Quebec at that time. Until the 1960s the award was accompanied with a medal hung from a ribbon, but this was changed in 1964. The Quebec order was renamed Ordre national du Mérite agricole in 2001, and its purpose remains to recognize excellence in agriculture. The award insignia is now a circular medal not suited for wear. While it is a prestigious award, it is not associated with the Ordre national du Québec and is not part of the broader Canadian honours system.
The Ministry of Agriculture was not the only provincial body to latch onto the desire to reward exceptional service. The Quebec government founded an Order of Scholastic Merit on March 22, 1928, through an act of the Legislative Assembly of Quebec. The order was divided into three levels: the Order of Scholastic Merit First Class (and the diploma of merit), the Order of Scholastic Merit Second Class (and the diploma of great merit), and the Order of Scholastic Merit Third Class (and the diploma of distinguished merit).
Teachers, school inspectors, and principals were all eligible. Those working in the Roman Catholic school system were nominated by a board composed of five members of the Roman Catholic Committee of Education, while those working in the Protestant school system were nominated by a board composed of five members of the Protestant Committee of Education. The superintendent of education was an ex officio member of both selection boards, as well as an ex officio member of the Order of Scholastic Merit Third Class. Each committee was allowed to define the criteria for award of the Order of Scholastic Merit. It is also interesting to note that each committee was responsible for designing its own insignia. Thus, the Roman Catholic insignia was different from the Protestant version. Awards of the Order of Scholastic Merit appear to have ceased during the Quiet Revolution in Quebec in the 1960s.
Prior to the founding of the Canadian honours system in 1967, British Columbia was the only other Canadian province to have a provincial order. The Order of the Dogwood was founded in 1966. It was British Columbia's highest award, and appointments were recommended to the lieutenant governor of British Columbia by a special advisory council. The order was awarded to British Columbians who had "served with the greatest distinction and excelled in any field of endeavour benefitting the people of the province of British Columbia or elsewhere." Between 1966 and 1989, the order was awarded a total of thirteen times, but it had largely fallen out of use by the late 1970s.
Modern Provincial Honours
In 1938, Prime Minster William Lyon Mackenzie King's deputy minister of justice warned him that the provinces were quite capable of creating their own awards and having their respective lieutenant governors sanction them. Although such awards would not be "official," there was a fear that, with time, they would become viewed as such. The incorporation of the provincial orders into the Canadian Order of Precedence — their acceptance as legitimate awards — was in many ways a concession to the provinces and came about partly out of a desire to avoid federal-provincial conflict.
Throughout the 1970s and 1980s, various lieutenant governors lobbied to be appointed ex officio members of the Order of Canada. Consideration had been given to making each lieutenant governor a vice-chancellor of the Order of Canada, but the Advisory Council of the Order of Canada did not favour this idea and it was abandoned.
The provincial orders have an interesting relationship with the Order of Canada: on average, 40 percent of those who receive a provincial honour are already recipients of one of the levels of the Order of Canada.
All but one of the modern provincial orders have been established by acts of the respective provincial legislatures and existed in provincial law prior to being recognized by the federal government.[2] The Queen is not the Sovereign of any of the provincial orders, though in some cases she has given approval for the use of the Royal Crown on insignia.
The provincial orders are included in the Canadian Order of Precedence not in the order they were created but rather in the order that each province requested the federal government to recognize its order. Although Alberta was the first province to create a provincial order, Quebec was the first — in 1984 — to request that its order be incorporated into the Canadian Order of Precedence and be recognized as an official award.
On May 9, 1991, the federal government passed Order-in-Council 1991-841, which recognized the provincial orders that existed at that time: the Ordre national du Québec, the Saskatchewan Order of Merit, the Order of Ontario, the Order of British Columbia, and the Alberta Order of Excellence. Since that time other provincial orders have been added so that each province now has one.
With the exception of Quebec, all of the provincial orders consist of one level; Quebec's has three. Generally, the orders are similar in that their criteria and the method used to select recipients are based on those used by the Order of Canada, with a strong emphasis on public nominations. The lieutenant governor of each province is the Chancellor of his or her respective order, except in Quebec where there is no Chancellor.
Each provincial order is accompanied by a full-size insignia; with the exception of Alberta and Prince Edward Island, all contain or are based on the respective provincial flowers. Recipients also receive a lapel pin and certificate. Insignia of nearly all the provincial orders are worn around the neck — or on a bow by women. The exception is the insignia of the Chevalier of the Ordre national du Québec, which is worn on the left breast.
In all provinces aside from Quebec, provincial and federal politicians and judges are prohibited from receiving a provincial order while in office. Saskatchewan, British Columbia, and Nova Scotia have a special Chancellor Chain worn by the lieutenant governor.
The Ordre national du Québec
Grand Officer of the Ordre national du Québec.
Officer of the Ordre national du Québec.
Origins: The Ordre national du Québec was established by an act of the Quebec National Assembly and was given Royal Assent on June 20, 1984. The order consists of three levels — Grand Officer, Officer, and Knight — and is therefore based on the Order of Canada as well as the more antique Quebec Order of Agricultural Merit. The levels of the Ordre national du Québec are based on the European model where Knights represent the lowest grade in an order. There is no Chancellor of the order, though in all but name the premier of Quebec acts in this capacity. The Ordre national du Québec is the most liberally awarded of the provincial orders.
Criteria: The order recognizes those men and women of Quebec who have distinguished themselves in science, technology, the arts, social action at home and abroad, or through accomplishments that are admired by all. The levels awarded to recipients are associated with the calibre of their contributions.
Appointments are made by the premier of Quebec on the advice of the Conseil de l'Ordre through an order-in-council. This committee is made up of nine members, all of whom are elected from within the order.
Nominations are made by the general public. The order may also be bestowed posthumously, making it one of the first in the Commonwealth to allow such appointments. There are no annual limits on the number of appointments to each level. However, there are usually three GOQ appointments, ten OQ appointments, and fifteen CQ appointments each year.
Insignia: The insignia varies according to the level, but all three have a feature in common: a stylized Geneva cross bisected by a shape representing the course of the St. Lawrence River, with a fleur-de-lys at the mouth of the river.
Knight of the Ordre national du Québec obverse.
Knight of the Ordre national du Québec reverse.
* • Grand Officer/grand officier/grande officière: An 18-carat gold Geneva cross measuring 60 mm wide with a white enamelled fleur-de-lys in the lower left-hand corner of the cross.
* • Officer/officier/officière: A Geneva cross of 18-carat gold and oxidized silver measuring 40 mm wide with a gold fleur-de-lys in the lower left-hand corner.
* • Knights/chevalier/chevalière: A circular silver medal 40 mm in diameter superimposed with a Geneva cross bisected by the St. Lawrence River and with a gold fleur-de-lys in the lower left-hand corner of the cross.
The reverse of all three insignia bears the motto of the order — HONNEUR AU PEUPLE DU QUÉBEC — as well as a three-digit serial number.
Suspender: For the insignia of the Grand Officers and Officers, a small loop is affixed to the middle of the uppermost part of the cross. Through this passes a 20 mm loop and the ribbon of the order. The insignia of a Knight is suspended by a small ball mounted on the top of the medal through which passes a ring and the ribbon of the order. The Knight's insignia is worn on the left breast.
Ribbon: The ribbon is 38 mm wide with a 12 mm central white stripe flanked by stripes of blue.
Postnominals: GOQ, OQ, CQ.
Lapel Badge: A small Geneva cross in gold with a white enamelled fleur-de-lys for GOQ; a small Geneva cross in oxidized silver with a gold fleur-de-lys for OQ; and a small silver disc bearing a Geneva cross in oxidized silver with a gold fleur-de-lys for CQ.
Other: The order adopted a motto on April 3, 1985: Honneur au peuple du Québec/"Honour to the People of Quebec." The insignia was designed by Madeleine Dansereau.
Number of Appointments: Grand Officer, 102; Officer, 321; Knight, 512.
The Saskatchewan Order of Merit
Saskatchewan Order of Merit 1st type obverse (1985–2005).
Saskatchewan Order of Merit 1st type ribbon (1995–2005).
Origins: The idea to create the Saskatchewan Order of Merit emerged in 1980 when the province marked its seventy-fifth anniversary as part of Confederation. At that time a special interdepartmental committee was founded to investigate the possibility of creating a provincial award.
The Saskatchewan Award of Merit was established by an act of the provincial legislature on June 5, 1985. The award was renamed the Saskatchewan Order of Merit by a subsequent act that received Royal Assent on May 10, 1988. The order consists of one level. Since 1988 the lieutenant governor of the province has served as Chancellor, and in 2001 the lieutenant governor was made an ex officio member. There is no annual maximum number of appointments to the order, though the number is usually fewer than ten.
In 2000 a provision was added to allow for posthumous awards (to be made within a year of the person's death). In 2001 a further provision was added to allow for honorary appointments, the first of which was His Royal Highness The Prince of Wales (Prince Charles).
Saskatchewan Order of Merit 2nd type obverse.
Saskatchewan Order of Merit 2nd type ribbon.
Criteria: The order is awarded in recognition of individual excellence, outstanding achievement, and exceptional contributions to the social, cultural, and economic well-being of the province and its residents. Appointments are made by the lieutenant governor on the advice of the Saskatchewan Order of Merit Advisory Council. The council consists of a person appointed by the president of the Executive Council (the premier of Saskatchewan), the chief justice of Saskatchewan (or the chief justice of the Queen's Bench), the provincial Cabinet secretary (or the clerk of the Executive Council), the president of one of the province's two major universities (the Universities of Saskatchewan and Regina), the provincial archivist, and not more than five members appointed by the president of the Executive Council.
Insignia: A 45 mm silver-gilt insignia in the shape of a six-pointed star (red lily). In the centre is the provincial shield of arms surmounted by a Royal Crown. The reverse is plain and contains a three-digit serial number. In 2005 the insignia of the order was changed to include the provincial motto MULTIS E GENTIBUS VIRES ("From many peoples, strength") in a red circlet surrounding the shield.
Suspender: A small ring is affixed to the uppermost arm of the red lily through which passes a 21 mm hanger and the ribbon of the order.
Ribbon: A 38 mm wide ribbon green in colour bordered on each side by 13 mm of gold. Prior to 2005 the ribbon was 38 mm wide with a central gold stripe bordered on either side by 13 mm of green. The change was made to help differentiate the ribbon from that used by the Saskatchewan Volunteer Medal.
Lapel Badge: A miniature version of the full-size insignia with a crown in the centre.
Postnominals: SOM.
Other: The insignia was designed by Vic Sotropa of the Saskatchewan Visual Identity Office.
Number of Appointments: 200.
The Order of Ontario
Order of Ontario.
Origins: On June 14, 1973, the Province of Ontario established the Ontario Medal for Good Citizenship to recognize outstanding citizens of Ontario. From this medal flowed the idea that Ontario should create its own order. Established by an act of the provincial parliament, the Order of Ontario received Royal Assent in 1986. The order consists of one level. The lieutenant governor is Chancellor of the order and an ex officio member. There is no annual maximum number of appointments, though the largest number appointed in a single year has been twenty-seven.
Criteria: The order is awarded to recognize the highest level of individual excellence and achievement in any field. Appointments are made by the lieutenant governor on the advice of the Advisory Council of the Order of Ontario, which consists of the chief justice of Ontario, the speaker of the Legislative Assembly of Ontario, the secretary of the provincial Cabinet, and a maximum of six distinguished members of the Order of Ontario.
Insignia: A stylized silver-gilt trillium 54 mm wide, enamelled in white, and superimposed on a triangular green enamel trillium leaf. In the centre is the provincial shield surmounted by a Royal Crown. The reverse is plain and contains a three-digit serial number.
Suspender: A small loop is affixed to the uppermost point of the trillium leaf through which passes a 21 mm loop and the ribbon of the order.
Ribbon: A 38 mm ribbon with outer edges of red and a central white stripe superimposed with stripes of gold, green, and gold.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: OOnt.
Other: The insignia of the order was designed by Bruce Beatty.
Number of Appointments: 597.
The Order of British Columbia
Order of British Columbia.
Origins: Established by an act of the provincial legislative assembly, the Order of British Columbia received Royal Assent on April 21, 1989. The order consists of one level. The lieutenant governor is Chancellor of the order and an ex officio member.
Criteria: The order is awarded to recognize those who have served with the greatest distinction and who have excelled in any field to the benefit of the people of the province or elsewhere. Appointments are made by the lieutenant governor through a decision of the lieutenant governor-in-council (the provincial Cabinet) on the recommendation of the Advisory Council. The Advisory Council of the Order of British Columbia consists of the lieutenant governor, the chief justice of British Columbia, the speaker of the Legislative Assembly of British Columbia, the president of one of British Columbia's public universities, the president of the Union of British Columbia Municipalities, the deputy minister responsible for the Order of British Columbia, and two recipients of the order from the previous year. Nominations are sought from the general public.
Insignia: A silver-gilt insignia 54 mm wide in the shape of a dogwood flower in white enamel with green leaves in the background. The centre of the flower is defaced with the provincial shield of arms surmounted by a Royal Crown. The reverse of the insignia is plain and contains a three-digit serial number.
Suspender: A small ring is mounted on the top of the insignia through which passes a 21 mm loop, through which passes the ribbon of the order.
Ribbon: A 38 mm wide ribbon composed of green, white, yellow, and dark blue.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: OBC.
Other: The insignia of the order was designed by Bruce Beatty.
Number of Appointments: 385.
The Alberta Order of Excellence
Alberta Order of Excellence.
Origins: Founded by an act of the provincial legislature, the Alberta Order of Excellence received Royal Assent on November 16, 1979, and is the oldest provincial order, though it was not awarded until 1981. The order started off as a badge/table medal not intended for wear but was later modified for wear around the neck. It consists of one level. The lieutenant governor is Chancellor of the order.
Criteria: The order is awarded to persons who have rendered service of the greatest distinction and of singular excellence for or on behalf of the residents of Alberta. It is awarded by the lieutenant governor on the advice of the Alberta Order of Excellence Council, which is made up of the lieutenant governor and six other Albertans who are chosen by the lieutenant governor-in-council (the provincial Cabinet). There is no limit on the number of annual awards; the largest number of appointments in a single year has been six.
Insignia: A silver-gilt cross 51 mm wide with four arms. Between each arm of the cross is a rose with four leaves. The arms are enamelled in translucent blue. The centre of the cross contains the arms of the province on a red background surrounded by the words THE ALBERTA ORDER OF EXCELLENCE on a white circlet. The reverse is plain, aside from a maple leaf and a wreath of wheat.
Suspender: A small ring is mounted on top of the cross through which passes a loop and the ribbon.
Ribbon: A 38 mm ribbon composed of blue, white, burgundy, and gold.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: AOE.
Number of Appointments: 140.
The Order of Prince Edward Island
Order of Prince Edward Island.
Origins: The Order of Prince Edward Island was founded by an act of the provincial legislature and received Royal Assent on May 29, 1997. The lieutenant governor of the province is chancellor of the order and is also an ex officio member. The order consists of one level, and a maximum of three awards may be made each year.
Criteria: The order is awarded in recognition of outstanding individual excellence and achievement and outstanding contributions to the social, cultural, or economic well-being of Prince Edward Island and its residents. It is awarded by the lieutenant governor on the advice of the Order of Prince Edward Island Advisory Council. Nominations come from the general public. The council is composed of the chief justice of Prince Edward Island, the president of the University of Prince Edward Island, the clerk of the Executive Council, and two residents of each county in the province (appointed by the lieutenant governor-in-council, i.e., the provincial Cabinet).
Insignia: A circular gold medal with the provincial shield in the centre circumscribed by the words MERIT — PRINCE EDWARD ISLAND on English issues and MÉRITE — ÎLE DU PRINCE ÉDOUARD on French issues. The reverse is engraved with a three-digit number starting at 01 for French issues, 001 for English issues, and C-1 for the Chancellor's insignia.
Suspender: A gold loop 21 mm in length.
Ribbon: A ribbon 38 mm wide edged with 3 mm of white, with three central stripes of white (6 mm) bordered by rust red (13 mm) on the left and green (13 mm) on the right.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: OPEI.
Number of Appointments: 60 .
The Order of Manitoba
Order of Manitoba.
Origins: The Order of Manitoba was founded by an act of the provincial legislature and was given Royal Assent on July 14, 1999. The order replaced the unofficial "Order of the Buffalo Hunt." It consists of one level, and the lieutenant governor of the province is the Chancellor of the order and an ex officio member. A maximum of eight appointments may be made annually (a special provision was made for twenty to be appointed in the year the order was founded).
Criteria: The order recognizes those who have demonstrated excellence and achievement in any field of endeavour that benefits in an outstanding manner the social, cultural, or economic well-being of Manitoba and its residents. Awards are made by the lieutenant governor on the advice of the Advisory Council of the Order of Manitoba, which consists of the chief justice of Manitoba, the clerk of the Executive Council, the president of one of the province's three universities, and not more than four people appointed by the lieutenant governor-in-council (the provincial Cabinet) for a term of three years. Any citizen may make a nomination.
Insignia: A stylized crocus (the provincial flower) 58 mm wide and enamelled in powder blue with a gold frame. In the centre is the provincial shield of arms surmounted by a Royal Crown. The reverse is plain and contains a three-digit serial number.
Suspender: A gold loop 21 mm in length.
Ribbon: A white ribbon 38 mm wide superimposed with two broad stripes of powder blue near the edges and a thin central stripe of red.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: OM (not to be confused with the Order of Merit).
Number of Appointments: 197.
The Order of New Brunswick
Order of New Brunswick.
Origins: The Order of New Brunswick was created by an act of the provincial legislature and was given Royal Assent on December 20, 2000. The design of the order's insignia was officially unveiled on November 19, 2001. The order consists of one level. The lieutenant governor of the province is Chancellor of the order and an ex officio member.
Criteria: The Order of New Brunswick recognizes those who have demonstrated excellence and achievement and who have made outstanding contributions to the social, cultural, or economic well-being of New Brunswick and its residents. The order is awarded by the lieutenant governor on the advice of the Advisory Council of the Order of New Brunswick, which consists of a chair (appointed by the provincial Cabinet), the chief justice of New Brunswick, the clerk of the Executive Council, the president of one of the province's universities, and three to five members of the community (appointed by the Executive Council, i.e., the provincial Cabinet). A maximum of ten appointments may be made each year. Nominations are sought from the general public.
Insignia: A stylized violet enamelled in deep blue with a silver-gilt frame. In the centre is the provincial shield of arms surmounted by a Royal Crown. The enamel was originally supposed to be deep purple (the purple violet is New Brunswick's provincial flower), but this proved too difficult to manufacture, so it was changed to deep blue. The reverse of the insignia bears a three-digit number.
Suspender: A gold loop 21 mm in length.
Ribbon: A 38 mm ribbon edged in red with a central yellow stripe superimposed with a purple stripe.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: ONB.
Number of Appointments: 131.
The Order of Nova Scotia
Order of Nova Scotia.
Origins: The Order of Nova Scotia was established by an act of the provincial legislature and was given Royal Assent on June 1, 2001. The order consists of one level. The lieutenant governor is Chancellor of the Order and an ex officio member. Nova Scotia has a curious history in terms of honours. In 1624, King James I created the Baronets of Nova Scotia (see chapter 4) primarily as a tool to raise funds for the colonization of Nova Scotia. Eventually, the baronetcies came to be awarded for merit. The insignia of the Baronets of Nova Scotia bears some resemblance to that of the Order of Nova Scotia in that the coat of arms used on each insignia is the same.
Criteria: The order is awarded to residents of Nova Scotia who have distinguished themselves by an outstanding contribution to the cultural life or to the social or economic well-being of the province, by an outstanding achievement, or by excelling in any field of endeavour to the benefit of the people of the province or elsewhere. Appointments are made by the lieutenant governor on the advice of the Advisory Council of the Order of Nova Scotia, which consists of a chair (appointed by the premier), the chief justice of Nova Scotia, the clerk of the Executive Council, three members appointed by the premier (one of whom is a president of one of the province's universities), a member appointed by the leader of Her Majesty's Loyal Opposition, and a member appointed by the leader of each recognized provincial political party. The order is unusual in that it may be awarded posthumously. A maximum of five appointments may be made each year (a special provision allowed for ten appointments in the year of the order's founding). Nominations are sought from the general public. At the time that the order was created all living former lieutenant governors of the province were appointed to the order.
Insignia: A 61 mm badge in the shape of a mayflower (the provincial flower) enamelled in white with a slight tinge of red and defaced with the provincial arms surmounted by a Royal Crown. The reverse of the insignia bears a three-digit number.
Suspender: A gold loop 21 mm in length.
Ribbon: A 32 mm ribbon of blue, white, gold, and red.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: ONS.
Other: The insignia was designed by Christopher Cairns.
Number of Appointments: 81 .
The Order of Newfoundland and Labrador
Order of Newfoundland and Labrador.
Origins: The Order of Newfoundland and Labrador was founded by an act of the provincial legislative assembly and was given Royal Assent on May 24, 2001. The concept for the order was initially developed in the late 1990s. The order consists of one level. The lieutenant governor of the province is Chancellor of the order as well as an ex officio member.
Criteria: The order is awarded for outstanding achievement or contribution in any field. Residents and former long-term residents of the Province of Newfoundland and Labrador are eligible for the order. A maximum of eight people may be appointed to the order per year. Appointments are made by the lieutenant governor on the advice of the Advisory Council of the Order of Newfoundland and Labrador, which is made up of the clerk of the Executive Council, two members of the order, and up to four other members appointed by the lieutenant governor-in-council (the provincial Cabinet).[3] Nominations are sought from the general public.
Insignia: The insignia, 54 mm wide, is a stylized rendering of the pitcher plant (the provincial floral emblem). The dark grey petals are made of labradorite (calcium sodium aluminum silicate), the provincial mineral, and are edged in gold. This is defaced with the arms of Newfoundland and Labrador surmounted by a Royal Crown. It is one of the few orders in the world whose insignia is made from stone. A three-digit number is engraved on the reverse.
Suspender: A gold loop 21 mm in length.
Ribbon: The ribbon is 35 mm and edged with 5 mm of green. A 5 mm central blue stripe is bordered on either side by 4 mm of white, 2 mm of gold, and an additional 4 mm of white.
Lapel Badge: A miniature version of the full-size insignia.
Postnominals: ONL.
Number of Appointments: 65 .
# 18
Military Valour Decorations
There is perhaps no more famous military decoration than the Victoria Cross. It has an almost universal recognition as the pre-eminent valour decoration. It also has elegantly simple criteria: "For most conspicuous bravery or some daring or pre-eminent act of valour or self-sacrifice or extreme devotion to duty in the presence of the enemy."[1]
The Victoria Cross was instituted by Queen Victoria on February 5, 1856 (see chapter 5). The designer remains unknown, though it is likely that the work was done by someone at Hancocks & Company of London,[2] and the original Victoria Cross prototype was certainly made of sterling silver by Hancocks. The appearance was deemed unsatisfactory, however, and bronze was procured. In fact, a very symbolic — and large — piece of bronze was used from a pair of bronze cannon that had been captured from the Russians in Sebastapol during the final battle of the Crimean War.
The quality of this bronze, which was Chinese in origin, was so inconsistent that the Victoria Cross had to be sand-cast rather than die-struck as most medals were. The sand-casting technique is used to this day. The recipient's details are hand-engraved on the reverse, and the cross is fitted into a special presentation case. Occasionally, the citation for which the award is being made is printed on the silk lining of the case's interior.
Canada's Military Valour Decorations find their origins not only in a variety of British gallantry awards but are also very closely related to the Canadian Bravery Decorations instituted in 1972. All three decorations — the Victoria Cross (VC), the Star of Military Valour (SMV), and the Medal of Military Valour (MMV) — are awarded for acts of valour in the presence of the enemy, the term enemy being defined broadly enough to include mutineers, armed rebels, and pirates. Valour is defined as an exemplary action demonstrated in the presence of the enemy, while bravery is an exemplary action demonstrated in a more general sense. Both require a significant level of personal risk.
Under the British system of honours, decorations for military valour included, among others, the Victoria Cross, the Distinguished Service Cross, the Military Cross, the Distinguished Flying Cross, the Distinguished Conduct Medal, the Military Medal, the Distinguished Service Medal, and the Distinguished Flying Medal. These awards were regularly bestowed upon Canadians during the First and Second World Wars as well as the Korean War. Between the Korean War and the Gulf War of 1990–91, no military valour awards were granted to members of the Royal Canadian Navy, the Canadian Army, or the Royal Canadian Air Force, though a number of bravery decorations and Meritorious Service Decorations were conferred. The reason was that Canada was not involved in any armed conflict during that interval, and there was therefore little need for a separate series of Military Valour Decorations. Meanwhile, the broader bravery decorations instituted in 1972 were regularly awarded to members of the Canadian Armed Forces.
Almost immediately after the establishment of the bravery decorations in 1972, questions were asked regarding the status of the Victoria Cross and whether or not Canadians would continue to be awarded it. Prime Minister Pierre Trudeau regularly dodged the question, always commenting that Canadians should receive "Canadian" decorations, but never stating definitively that the Victoria Cross was never to be awarded to a Canadian again. The country was, after all, not at war.
Throughout the late 1980s and early 1990s, veterans' groups — most notably the Royal Canadian Legion — and other civic-minded organizations lobbied the government for the reinstatement of the Victoria Cross as Canada's premier decoration for military valour. This cause was supported by all parties in Parliament at the time, most notably the New Democratic Party, which was unanimously in favour. In 1987, Prime Minister Brian Mulroney assured members of the Royal Canadian Legion that the VC would become "the cornerstone of Canada's Military Valour Decorations."
The question of creating a series of military valour awards had long been a topic of discussion for the Honours Policy Committee. In late 1990 the committee developed a proposal to establish a series of three Military Valour Awards. It largely ignored the various calls for the reinstatement of the VC and instead proposed that three new Military Valour Decorations be created: the Cross of Military Valour, the Star of Military Valour, and the Medal of Military Valour.
Conveniently, in February 1991, a press leak resulted in a front-page article in the Globe and Mail that forced the Mulroney government to reconsider its decision. By October of that same year, Saskatchewan MP Doug Fee had introduced a private member's bill, the Victoria Cross Act, which sought to reinstate the decoration. After considerable lobbying by Fee, the bill received support from all parties in the House of Commons, and no member spoke against it.
At the direction of the prime minister the Honours Policy Committee re-examined the problem and decided to reinstate the Victoria Cross. At the same time two new decorations were to be established: the Star of Military Valour and the Medal of Military Valour. These were based on the Star of Courage and the Medal of Bravery respectively.
Such proposals for specific valour decorations for the military were not new. In 1966, shortly before the question of creating the Order of Canada came before Cabinet, the Department of National Defence had developed a proposal for the establishment of a "Cross of Courage" and a "Medal for Bravery."[3]
In the spring of 1993 the Cabinet agreed to the creation of the Canadian VC, the SMV, and the MMV, which were subsequently approved by the Queen on February 2, 1993. Award of the decorations was made retroactive to January 1, 1993.
Recipients of the awards are selected by the Military Valour Decorations Advisory Committee, which consists of six people: one appointed by the governor general and five appointed by the chief of the defence staff. The committee submits its recommendations to the chief of the defence staff, who then transmits them, upon approval, to the governor general. Under certain circumstances the governor general may delegate the power to award any of the Military Valour Decorations to field commanders, though their decisions are subject to the governor general's approval. Only members of the Canadian Armed Forces and allied forces are eligible for the Military Valour Decorations. Awards can also be cancelled or revoked by the governor general.
There are only a few differences between the original Victoria Cross and the Canadian Victoria Cross, the main one being that the Canadian cross can be revoked, whereas the British award, once bestowed, can never be rescinded. The term enemy is much more broadly defined for the Canadian VC as "a hostile armed force [including] armed mutineers, armed rebels, armed rioters and armed pirates." Canada therefore does not have to be in a formal state of war for members of the forces to be eligible for the Canadian VC or other Military Valour Decorations.
Canada is not the first country to duplicate the original VC. The government of New Zealand established the New Zealand Cross on March 10, 1869, by an order-in-council of the governor. This served as a colonial version of the Victoria Cross. However, it did not represent some great stride toward national independence in the realm of honours but was rather a result of the fact that colonial troops were ineligible to receive the Victoria Cross at that time, and the New Zealand authorities wished to reward gallant soldiers who had served in the Maori Wars (1860–72). The New Zealand Cross was phased out in 1910, though a modified version was reintroduced in 1999 to serve as the New Zealand equivalent of the George Cross. The most ignominious attempt to duplicate the VC was made by the Ugandan dictator Idi Amin, who instituted a "Victorious Cross" that was practically identical in design and criteria to the VC, with the exception that President Amin's effigy replaced the lion and crown on the VC's obverse. Not surprisingly, Amin himself was one of the recipients!
Australia instituted the Australian Victoria Cross in January 1991, two years before the Canadian VC was established. Indeed, actions by the Australian government to create the Australian Victoria Cross and the Australian Star of Gallantry and Medal for Gallantry helped to spur the development of the Canadian Military Valour Decorations. Whereas Australia had used the Order of Canada, the Cross of Valour, the Star of Courage, and the Medal of Bravery as the basic structure for its system of honours established in 1976, Canada in turn borrowed Australia's reinstatement of the Victoria Cross and two other gallantry awards as the basis for its own. As noted above, New Zealand instituted its own Victoria Cross in 1999.
The Canadian Victoria Cross is patterned directly after the original Victoria Cross. Its slightly modified design was drawn by Bruce Beatty, with later technical drawings executed by Cathy Bursey-Sabourin, Fraser Herald at the Canadian Heraldic Authority. The Star of Military Valour and the Medal of Military Valour were also designed by Beatty.
The first award of the Military Valour Decorations occurred on October 25, 2006, when Sergeant Patrick Tower was awarded the Star of Military Valour and Sergeant Michael Denine, Master Corporal Colin Fitzgerald, and Private Jason Lamont were awarded the Medal of Military Valour for their heroism while serving in Afghanistan.
The Victoria Cross (Canada)
Victoria Cross obverse.
The story behind the production of the Canadian Victoria Cross provides an interesting window into the symbolic importance that the VC has. Great attention to detail was invested in creating the Canadian Victoria Cross while respecting the splendid patrimony of the original Victoria Cross and its recipients from throughout the Commonwealth.
After the establishment of the Military Valour Decorations in 1993, there was no rush to have them produced; indeed, it was not until 2001 that the Star of Military Valour and the Medal of Military Valour were struck. The Victoria Cross was quite a different item because it could not be struck; it had to be cast. The entire project was left in abeyance for more than a decade. Canada was not involved in any large-scale armed conflict, and to some degree federal officials were lulled into the false sense that members of the Canadian Armed Forces serving overseas were never faced with the sort of armed conflict that could warrant the awarding of the Military Valour Decorations.
At various times officials discussed the question of who should manufacture the Canadian VC and from what material it should be fashioned. Initially, it was going to be put out to public tender, with the lowest-cost bidder taking the contract, but thankfully that fate did not befall the Canadian VC. Nevertheless, discussion about what the VC should be made of continued. A senior official had the idea of making the Canadian VC out of an artillery piece used at the Battle of Vimy Ridge until it was pointed out that only one known artillery piece from that conflict remained and such an important piece of military history could hardly be destroyed even to make the VC. Another suggestion was put forward that the VC should be made from one of the original rails that made up the transcontinental railway that united Canada in 1885. This idea was initially well received until it was pointed out that a medal could not possibly be cast out of the sort of steel used to make railway tracks. There was also a symbolic reason that doomed this proposal: the fact that most of the steel rails utilized to create the Canadian Pacific Railway were manufactured by the steel producer Krupp, the same company that made most of the armaments for Germany in the First and Second World Wars.
Always in the background was the possibility of contracting Hancocks & Company to cast the Canadian Victoria Crosses from the same metal used for the British, Australian, and New Zealand VCs. This idea was ruled out as a possibility because the VC had to be made in Canada. One frustrated official from the Department of National Defence commented, "If you want to make it so 'Canadian,' I suppose it will have to be carved from a piece of moose antler dipped in maple syrup."
The initial idea for what to make the Canadian VC out of came from Major Carl Gauthier of the Department of National Defence. One afternoon in May 2005, Gauthier was visited by the author, and among other things, they discussed the perennial question of what to fashion the Canadian VC from. In the brainstorming session, Gauthier initially suggested that part of the original VC bronze be mixed with a remnant of bronze from a blown-up statue of Queen Victoria that was destroyed by the Front de liberation du Québec in the late 1960s. The bronze continued to be housed by a Canadian museum and was available. However, using an artifact "created" by a terrorist attack was, after some reflection, not palatable. While the connection to Queen Victoria was excellent, the circumstances under which the statute came to be destroyed were fraught with problems. Another option raised was using part of the original bell from the tower of the Centre Block of Parliament, which was destroyed by fire in 1916.
Victoria Cross reverse.
Bar for the Victoria Cross.
Victoria Cross lapel pin.
The author had recently purchased a bronze 1867 Confederation Medal that was sitting on Gauthier's desk. After a great deal of conversation, it was suggested that the original VC metal should be mixed with a Confederation Medal (see chapter 10). The symbolism was ideal — the combination of the original metal with the very first medal struck for Canada and awarded to Canadians. The obverse bears the effigy of Queen Victoria, originator and namesake of the Victoria Cross and the monarch who assented to the British North America Act that created modern Canada, while the reverse depicts an allegorical image of the four founding provinces and Britannia. To this would be added metal mined and found in Canada to add an indigenous element to the mix.
In 2006 the Victoria Cross Production and Planning Group was established to devise a plan to have the Canadian Victoria Cross manufactured. This group, led by the office of the secretary to the governor general, included representatives from the Department of National Defence, Veterans Affairs Canada, the Department of Natural Resources, and the Department of Canadian Heritage. Representatives from the Royal Canadian Mint joined partway through the committee's deliberations. The group decided that the composition of the medal for the VC should be made up of a Confederation Medal, metallic elements from across Canada, and a piece of bronze from the original VC. The government of the United Kingdom donated a piece of the bronze used in the British VC for the Canadian project. Scientists from Natural Resources Canada made an analysis of the composition of a number of the VCs held by the Canadian War Museum to determine the metallurgical makeup in an effort to create a precise formula for the metal.
As with the original VC, the Canadian VC was cast and not struck. This required the development of three-dimensional artwork by the Royal Canadian Mint and the Canadian Heraldic Authority. From this work dies were produced with the obverse and reverse of the VC in negative and from these wax positives were made from which the VC was cast. Trees of these VC wax positives were created and then immersed in a ceramic mixture. Once the ceramic hardened it was heated to remove the wax, leaving behind the highly detailed VC moulds within the ceramic.
In December 2006 the casting process was initiated at the Materials Technology Laboratory of Natural Resources Canada in Ottawa. First the mixture of metal was combined to make seven ingots of Canadian VC bronze called the Rideau alloy. Following the preparation of the ingots, a number were melted for the production of the actual VCs, with the molten bronze being poured into the ceramic moulds, each resulting in four rough VCs. The incomplete insignia were then sent to the Royal Canadian Mint for hand-finishing and patination. The bronze remnants were subsequently melted together into another spare ingot.
The first two specimen VCs were sent to Her Majesty the Queen in January 2007, and the formal unveiling of the Canadian VC took place at Rideau Hall in May 2008.
The Canadian VC is housed in a brown leatherette case lined with silk and cream felt. The outside lid is edged with gilt tooling, and the centre contains the text V.C. CANADA. The cases are identical to those issued with the British, Australian, and New Zealand VCs, aside from the inclusion of the text CANADA. Hancocks was contracted to produce the cases.
The actual insignia of the Canadian VC differs slightly from the original VC; instead of the motto FOR VALOUR that appears on the obverse of the original VC, the Canadian version bears the same saying in Latin: PRO VALORE. This was to avoid the problem of using a lengthy bilingual motto. The other difference is the inclusion of a fleur-de-lys in the scroll that bears the motto. The original VC contains only a thistle, shamrock, and rose, representing the component parts of the United Kingdom — Scotland, Ireland, and England. All of these symbols appear in the Royal Arms of Canada, as well, along with the fleur-de-lys of royalist France. The famous Canadian patriotic song of the South African War (1899–1902) and First World War, "The Maple Leaf Forever," contains a verse "With lily, thistle, shamrock, rose — the maple leaf forever."
Like the original VC, each individual Canadian VC contains a small secret marking known only to the Royal Canadian Mint and the Department of National Defence that verifies that it is an authentic VC.
The Canadian Victoria Cross was formally unveiled by Governor General Michaëlle Jean and Prime Minister Stephen Harper on May 16, 2008, in the ballroom at Rideau Hall. In attendance were a myriad of veterans of the Second World War and Korean War (many of whom had been decorated for bravery), many of the recipients of the Military Valour Decorations, members of the Canadian Armed Forces, cadets, and people associated with the project.
Criteria: Awarded for the most conspicuous bravery, a daring or pre-eminent act of valour or self-sacrifice, or extreme devotion to duty in the presence of the enemy.
Insignia: A bronze straight-armed cross pattée with raised edges measuring 38 mm across. The obverse depicts a lion guardant standing on the Royal Crown, and below the crown, a scroll bears the inscription PRO VALORE. The reverse is plain with raised edges.
Suspender: A small semicircle with a hole in the centre is located on the uppermost arm of the cross. Through this passes a small bronze link connected to a V below a straight bar ornamented with laurel leaves.
Ribbon: A crimson ribbon 38 mm wide. Recipients wear a single miniature of the VC on their undress ribbons. A second award is denoted by a second miniature VC worn on the undress ribbon.
Bars: Additional awards of the VC are signified by a bronze bar ornamented with laurel leaves.
Naming: Engraved capitals on the reverse of the suspender with the recipient's rank, initials, name, and unit, while the date of action is in the centre of the reverse.
Lapel Badge: A small bronze cross with a Royal Crown in the centre.
Postnominals: VC.
Other: Twenty VCs were created and eight bars were produced. In addition to this, eight specimens made of commercial bronze were generated for display in public institutions. These are engraved with an inventory number — S-1 to S-8 — with the year of manufacture 2007 on the reverse of the insignia while the reverse of the suspender is engraved with SPECIMEN.
Number of Awards: None.
The Star of Military Valour
Star of Military Valour obverse.
Star of Military Valour reverse.
Criteria: Awarded for distinguished and valiant service in the presence of the enemy.
Insignia: A four-pointed gold star with a maple leaf in each of the four angles. In the centre of the obverse is a gold maple leaf on a red enamelled field surrounded by a silver laurel wreath. The reverse is plain with the Royal Cypher and Royal Crown and PRO VALORE. Insignia are stamped on the lower arm of the reverse along with a maker's name: RIDEAU STG or CARAT STER:.
Star of Military Valour lapel pin.
Suspender: A small ball at the top of the star through which passes a ring. The Star of Military Valour is worn on the left breast.
Ribbon: A crimson ribbon 32 mm wide. Set 2 mm from either edge is a white stripe 5 mm wide.
Bars: A gold bar bearing a single maple leaf in the centre. A single gold maple leaf is worn in the centre of the undress ribbon for second awards.
Lapel Badge: A small gold version of the insignia.
Naming: Capitals, abbreviated rank, initials, and surname are engraved on the reverse below PRO VALORE.
Postnominals: SMV or ÉVM.
Other: A manufacturer's mark is impressed on the lower arm of the reverse.
Number: 20.
The Medal of Military Valour
Medal of Military Valour obverse.
Medal of Military Valour reverse.
Criteria: Awarded for an act of valour or devotion to duty in the presence of the enemy.
Insignia: The Medal of Military Valour is circular and made of gold. On the obverse is a single maple leaf surrounded by a laurel wreath. The reverse bears the Royal Cypher and Royal Crown in the centre surrounded by PRO VALORE at the base.
Medal of Military Valour lapel pin.
Suspender: A straight suspender with a fleur-de-lys in the centre mounted to the disc of the medal.
Ribbon: A crimson ribbon 32 mm wide with three 3 mm white stripes — one in the centre of the ribbon, the others set 2 mm from either edge of the ribbon.
Bars: A gold bar bearing a single maple leaf in the centre. A single gold maple leaf is worn on the undress ribbon for second awards.
Lapel Badge: A small silver version of the insignia.
Naming: Capitals, abbreviated rank, initials, and surname are engraved on the rim.
Postnominals: MMV or MVM.
Other: Those insignia struck by Rideau Ltée have a stamp on the edge of the suspender, while those made by Carat Médailles are stamped with CARAT STER on the reverse of the suspender.
Number: 89.
# 19
Decorations for Bravery
Prior to the establishment of the Order of Canada in 1967 and dating back to pre-Confederation times, Canadians were eligible for a variety of British bravery and gallantry awards. These included but were not limited to the Victoria Cross, the George Cross, the Albert Medal, the George Medal, the Queen's Commendation for Brave Conduct, and the various levels of the Order of the British Empire for gallantry (see chapter 5). Non-governmental organizations such as the Royal Life Saving Society of Canada and the Royal Canadian Humane Association have also long possessed separate awards for life-saving.
The earliest Canadian bravery award was the Government of Canada Medal "For Gallant and Humane Services Rendered in Life Saving from Shipwrecked Vessels." Established in 1868 by Minister of Marine and Fisheries Peter Mitchell — who was himself the owner of a shipping company and a former premier of New Brunswick — the medal was issued in gold and silver. The sparingly bestowed medal appears to have fallen into disuse by 1913; in total, only six gold and eleven silver medals were issued.[1] Although obscure, this medal was the first uniquely government-sanctioned Canadian decoration for bravery. The dominion government also occasionally presented engraved gold pocket watches in the 1890s, as well, though these appear to have been treated as ad hoc junior awards. As the broader British honours system grew to include more bravery decorations for civilians, the need for such Canadian awards temporarily diminished.
Royal Canadian Humane Association Bravery Medal in bronze large issue obverse.
During the years that Canada used the British honours system, Canadians became familiar with a number of awards. The George Medal, established in 1940 by King George VI, and the King's (or Queen's) Commendation for Brave Conduct became the primary forms of recognition accorded to Canadian civilians who displayed exemplary bravery. These awards were also bestowed upon members of the Royal Canadian Navy, the Canadian Army, and the Royal Canadian Air Force. Less frequently used in the Canadian context was the Order of the British Empire. Although it was primarily a military and civilian award for meritorious service, provisions had been made at the time of its establishment for the order to be used also as a junior bravery award. Even at times when all other British honours were prohibited, provisions were made for the continuance of bravery awards. Under the British honours system, Canadians received ten George Crosses,[2] seventy-seven George Medals, twenty-two British Empire Medals for Gallantry, and numerous Queen's Commendations for Brave Conduct.
The small number of British bravery and gallantry awards is somewhat deceiving. From 1894 until 2001, the Royal Canadian Humane Association (RCHA) awarded nearly six thousand bravery medals. (At various times a member of the Canadian Armed Forces was permitted to wear the RCHA Bravery Medal on his or her uniform, albeit on the right side of the chest.) In many ways the RCHA Bravery Medals were the first indi-genous Canadian awards for bravery.
The RCHA, which was modelled on the Royal Humane Association in Britain, awarded medals in gold, silver, and bronze. These bravery awards made up for the absence of a Canadian bravery or life-saving medal. The association would receive nominations from the general public, and a special committee would then decide on what awards were to be made. The medals were usually presented by the governor general or a lieutenant governor. In 2001 the RCHA ceased to operate; however, in 2003, it was reactivated and has resumed presenting awards for bravery.
Royal Canadian Humane Association Bravery Medal in bronze large issue reverse.
Canada's three bravery awards find their immediate origins in the George Cross, the George Medal, and the Medal of Courage of the Order of Canada. The Medal of Courage was originally intended to serve as a Canadian version of the George Medal, but Prime Minister Lester Pearson also wanted to continue to use the Victoria Cross and George Cross as part of the broader Canadian honours system. By March 1968, however, Pearson had reconsidered this arrangement and decided that the Canadian government should no longer make recommendations for Canadians to receive British bravery awards,[3] leaving the Medal of Courage as the only bravery award available to Canadians. It was suddenly required to replace not only the George Medal but also the Victoria Cross, the George Cross, and the Queen's Commendation for Brave Conduct, not to mention the myriad of service-specific bravery awards such as the Distinguished Service Cross, the Military Cross, and the Distinguished Flying Cross.
Royal Canadian Humane Association Bravery Medal in bronze small issue.
Within a short period of time "it was decided... that the Medal of Courage which was incorporated into the Order of Canada would not allow sufficiently for the recognition of acts of bravery."[4] A variety of solutions were considered. In the summer of 1968 the secretary to the governor general, Esmond Butler, suggested that the Medal of Courage be divided into three grades within the Order of Canada. This idea was quickly discarded, however, in the face of opposition from the Department of National Defence and the Royal Canadian Mounted Police.[5] It was agreed instead that "the Order of Canada should be persuaded to leave the field of bravery awards."[6] To this end, the Working Group on Honours — a subcommittee of the Decorations Committee — prepared a proposal for three separate bravery awards: a Hero's Cross, a Medal of Courage, and a Canadian Life-Saving Medal. All would be separate from the Order of Canada.[7] This proposal for three separate decorations emerged in October 1968, and over the next three years would be refined into the Cross of Valour, the Star of Courage, and the Medal of Bravery, though the exact names of each award were not to be devised until late 1971.
Although it took nearly four years to establish the Canadian bravery decorations, there was definitely a sense of urgency, since the backlog of bravery award nominations extended back to 1965. The last George Medals to be awarded to Canadians were presented in 1964 to Sergeant Joseph Lessard and Lieutenant-Colonel Paul Mayer, and the last appointment made to the Order of the British Empire for gallantry came in early 1968. After this the government ceased all recommendations for imperial gallantry decorations, including the Queen's Commendation for Brave Conduct.
Events aboard HMCS Kootenay in October 1969 helped to exped-ite the creation of the Canadian bravery decorations. The Kootenay suffered a catastrophic explosion in its engine room, and the ensuing fire threatened the entire ship. Through the selfless actions of a few, many lives — and the ship itself — were saved. The stories of Chief Warrant Officer Vaino Partanen and Sergeant Lewis Stringer — both of whom died on the Kootenay, Partanen while trying to report the situation and Stringer as a result of smoke inhalation he suffered while he cleared the galley — became widely known, as did the exemplary conduct of the remainder of the Kootenay's crew.
There were calls in the press for the men to be awarded the Medal of Courage, which was not practical since the government had already decided to cancel the Medal of Courage and replace it with a new series of decorations. The question of honours for the sailors persisted, however, and the subject was even brought up in Parliament. On December 30, 1971, D.W. Groos, the MP for Victoria, asked "would the PM] use his influence to speed [the awards] up as they are now more than two years overdue?"[[8] Prime Minister Trudeau told the House of Commons that the new system was under consideration, knowing full well that the CV, SC, and MB were in the final stages of preparation.
On July 28, 1971, the Cabinet met to discuss the proposed restructuring of the Canadian honours system, which included the creation of the three bravery decorations. The Cabinet did not spend much time on the bravery awards, being more fixated on the new structure of the Order of Canada and the brand-new Order of Military Merit. The Decorations Committee proposed that the highest level of bravery decoration should be called the Kootenay Cross. The second level of bravery decoration was initially given a variety of names: the Laurentian Star, the Star of the North, and the Cross of Courage. All of these ideas were poorly received, and while the overall proposal for three bravery decorations was approved, the name of the first and second degrees of the decoration were left "to be decided," while the third was designated the Medal for Bravery, or the MB.[9] The term bravery was chosen in place of gallantry because it translates into both official languages more fluidly.
The Queen approved the new bravery decorations on May 10, 1972, and they came into effect on July 1 of that year. The three decorations were designated the Cross of Valour, the Star of Courage, and the Medal of Bravery. The first awards were gazetted on July 28, 1972. Appropriately, the first Crosses of Valour, Stars of Courage, and Medals of Bravery were awarded to crewmen of HMCS Kootenay.
Recipients of the Canadian Bravery Decorations are chosen by the Canadian Decorations Advisory Committee. This committee consists of up to ten members: delegates of the clerk of the Privy Council, the secretary to the governor general, the deputy minister of national defence, the deputy minister of transport, the commissioner of the Royal Canadian Mounted Police, and up to four other members — normally a member of the Canadian Armed Forces and specialists in the fire, medical, and aviation fields — appointed by the governor general.
When Australia established its bravery decorations in 1975, it used the same structure — and almost the same criteria — as the Canadian Decorations for Bravery. The British also learned from the Canadian example: in 1974 they established the Queen's Gallantry Medal, which is junior to the George Medal and similar to the Medal of Bravery in terms of criteria.
All three decorations were designed by Bruce Beatty. When his original drawings were presented to the Decorations Committee for approval, the committee was pleased but insisted that the Cross of Valour, which had origin-ally been designed for wear on the breast, be changed to a neck decoration. This change was apparently made merely to make the decoration seem more "important." The drawings approved by the committee all utilized the same watered red ribbon, with no differentiation between them as the topic had not yet been discussed. As almost a last-minute augmentation, Beatty added two stripes of blue to the ribbon on the Star of Courage and three to that of the Medal of Bravery to differentiate them, blue just being the colour on his brush at the moment. When the committee approved the decorations, Beatty asked about the ribbon designs. The response was that those on the draft paintings were perfectly fine! Subsequently, all of the designs were approved by the Queen.
Recipients of the Cross of Valour, the Star of Courage, and the Medal of Bravery receive certificates bearing their names, citations of their awards, and the signatures of the governor general and the secretary to the governor general.
The status of the bravery decorations has remained largely unchanged since the establishment of the Military Valour Decorations (see chapter 18) in 1993. These new decorations are specifically intended to recognize only acts of valour demonstrated while under fire in the face of an enemy, while the bravery decorations cover all other instances in which people put their lives at risk to save or attempt to save someone.
The Cross of Valour
Cross of Valour obverse.
Cross of Valour prototype.
Criteria: Awarded only for acts of the most conspicuous courage in circumstances of extreme peril.
Insignia: A gold cross of four equal limbs. The obverse is enamelled in translucent red with machine work in a lozenge pattern below. The entire insignia is edged with gold. In the centre of the cross, a gold maple leaf is surrounded by a gold laurel wreath. The reverse is plain, with the Royal Cypher, Royal Crown, and VALOUR — VAILLANCE recessed into the insignia.
Suspender: At the top of the cross is a small ball through which passes a ring attached to a 20 mm straight suspender. The Cross of Valour is worn around the neck for men and on a bow for women.
Cross of Valour reverse.
Cross of Valour lapel pin.
Cross of Valour second award bar.
Ribbon: In light crimson and 38 mm wide. A recipient wears a miniature of the CV on the undress ribbon.
Bars: A gold maple leaf mounted to the 20 mm straight suspender (neck hanger). None has been awarded to date. A recipient of the bar would place two miniature CV insignias on the undress ribbon bar.
Naming: Capitals, initials, and surname are engraved on a single line on the reverse with the date of incident engraved below.
Lapel Badge: A small gold version of the insignia with red enamel.
Postnominals: CV.
Number: 20.
Other: When the Cross of Valour was sanctioned in 1972, twenty-five insignia were ordered from Garrard & Co.; the Chancellery is still using this original shipment. The lower arm on the reverse was stamped with the 18-carat hallmark. The edge of the lower arm is numbered.
The Star of Courage
Star of Courage obverse.
Star of Courage reverse.
Criteria: Awarded for acts of conspicuous courage in circumstances of great peril.
Insignia: A four-pointed silver star with a maple leaf in each of the four angles. In the centre of the obverse is a gold maple leaf surrounded by a gold laurel wreath. There are two reverse types; both display a representation of the Royal Cypher EIIR and a Royal Crown with the word COURAGE. However, on the first type these details are recessed, while on the second type — struck by Carat Médailles and Pressed Metal Products — these details are in bas relief.
Star of Courage lapel pin.
Suspender: A small ball at the top of the star through which passes a ring and the ribbon. Issues struck by Spink & Son and the Royal Canadian Mint have a 12 mm ring of heavy-gauge wire, while those by Carat Médailles and Rideau Ltée are 16 mm in diameter and a narrower gauge wire.
Ribbon: A light crimson ribbon 32 mm wide with two 5 mm blue stripes set 2 mm from either edge.
Bars: A gold bar bearing a single maple leaf in the centre. A single gold maple leaf is worn in the centre of the undress ribbon for second awards.
Naming: Capitals, initials, and surname are engraved on the reverse.
Lapel Badge: A small silver-and-gold version of the insignia.
Postnominals: SC or ÉC.
Other: The first Stars of Courage were manufactured by Spink & Son. Some of these are stamped STERLING on the reverse of the lower arm. By the early 1980s, the Royal Canadian Mint began striking them. In the mid-1980s, Rideau Ltée began production, and these are stamped with RIDEAU STG. Carat Médailles, in the early 2000s, began producing the insignia, which are stamped CARAT STER. Those issued by Birks are stamped BIRKS STER in two lines. Pressed Metal Products uses PMP SS.
Number: 447.
The Medal of Bravery
Medal of Bravery reverse.
Medal of Bravery obverse.
Criteria: Awarded for acts of bravery in hazardous circumstances.
Insignia: A circular silver medal. On the obverse is a single maple leaf surrounded by a laurel wreath. The reverse bears the Royal Cypher and Royal Crown in the centre surrounded by BRAVERY — BRAVOURE, with a six-leaf laurel flourish at the base.
Suspender: A straight suspender with a fleur-de-lys in the centre mounted to the disc of the medal.
Medal of Bravery second award bar.
Medal of Bravery lapel pin.
First type Medal of Bravery certificate.
Ribbon: In light crimson, 32 mm wide, with three 3 mm blue stripes — one in the centre of the ribbon, the other two set 2 mm from either edge of the ribbon.
Bars: A silver bar bearing a single maple leaf in the centre. A single silver maple leaf is worn on the undress ribbon for second awards.
Naming: Capitals, initials, and surname are engraved on the edge of the medal.
Lapel Badge: A small silver version of the insignia.
Postnominals: MB.
Other: Initially, the Medal of Bravery was made by Spink & Son. In the early 1980s the Royal Canadian Mint began striking the medal, after which the contract was awarded to Rideau Ltée, Birks, and Pressed Metal Products. The last named is stamped on the side of the suspender as PMP SS.
Number: 3,129.
# 20
The Meritorious Service Decorations
The Meritorious Service Decorations, which include the Meritorious Service Cross and the Meritorious Service Medal, serve as junior awards to the Order of Canada, the Order of Military Merit, and the Order of Merit of the Police Forces, which are typically used to recognize a lifetime of service. These decorations reward service or a specific achievement that falls short of that normally required for admission to one of the national orders. Essentially, the decorations recognize short-term merit while the orders are dedicated to long-term merit recognition. Whether one is awarded the cross or the medal is determined by the calibre of the specific achievement or service rendered.
The military division of the Meritorious Service Cross is similar to the Distinguished Service Order, which was awarded to Canadian officers from the South African War (1899–1902) until the Korean War. The civilian division compares with the Commander level of the Order of the British Empire. The military division of the Meritorious Service Medal, however, bears little relation to its British predecessor of the same name. The British Meritorious Service Medal was typically awarded to senior non-commissioned officers who had served more than twenty years in the armed forces. It was also infrequently used as a junior gallantry award. In contrast, both the military and civil divisions of the Canadian Meritorious Service Medal are awarded for deeds similar to that recognized by the Officer and Member level of the Order of the British Empire. Although considered intermediate awards, both the cross and the medal are awarded sparingly.
The origins of the Meritorious Service Decorations can be found in the Canada Medal, which was instituted in 1943 but was never awarded. The Canada Medal was initially intended to be a junior award for special service of a high degree of merit such as the discharge of special duties above and beyond the recipient's ordinary work, or highly meritorious performance of ordinary duties that entailed work of an especially trying nature or the display of a high degree of initiative and forethought.[1]
Prototype Meritorious Service Decoration.
Prototype Meritorious Service Medal.
Originally, the Canada Medal was supposed to be instituted along with a five-level Canadian Order — similar in structure to the Order of the British Empire — but when the government of the day refused to allow the creation of such an order, the Canada Medal was quickly elevated to a senior award that would be conferred upon foreign heads of state and corporals alike. Needless to say, this arrangement was entirely impractical, and though the medal remained an official award from 1943 until 1967, it was never awarded. Initially, the Canada Medal was to be called the Canadian Meritorious Award, and in 1944 there was a further proposal for a Meritorious Service Star.[2]
Shortly before the creation of the Order of Canada, Paul Hellyer, the minister of national defence, tabled a proposal in the Cabinet that called for the initiation of a comprehensive set of Canadian military awards. The proposal sought to establish two awards for meritorious service — the Forces Meritorious Decoration and the Forces Meritorious Medal — both of which were designed to recognize exceptionally meritorious service over a period of time or in a specific deed.[3] The entire proposal was shelved at the request of Prime Minister Lester Pearson, who was far more interested in getting his Cabinet to agree to the establishment of the Order of Canada and did not want to overload them with honours proposals.
After the establishment of the Order of Military Merit in 1972, members of the Canadian Armed Forces were, on the whole, happy with their new awards. At the most senior level, however, there was still a desire to create an award for meritorious service rendered in a specific deed. Neither the Order of Canada nor the Order of Military Merit could be used for this purpose,[4] and while the country had recently adopted a set of three bravery awards, there was still considered to be a significant gap in the honours system.
Throughout the late 1970s the Government Working Group on Honours examined the shortcomings of the Canadian honours system. Its report of July 1980 called for the creation of the Meritorious Service Cross, along with a variety of other additions.
On September 15, 1982, Prime Minister Pierre Trudeau approved the Canadian Armed Forces Honours Policy. The report indicated that there was a need for an intermediate award to be ranked below the Order of Military Merit to recognize "short-term outstanding achievement."[5] The military division of the Meritorious Service Cross was founded on June 11, 1984, when the Queen signed the letters patent.
The civil division of the Meritorious Service Cross and both divisions of the Meritorious Service Medal were created in part due to problems experienced by the Advisory Council of the Order of Canada. Since its inception, the council had complained that there was no junior honour to the Order of Canada. There had also been constant worries about the practice of awarding the Order of Canada in recognition of a single achievement, something that had been done in a number of instances, most notably for Olympic medallists. The addition of a civil Meritorious Service Cross and Meritorious Service Medal was seen by members of the Honours Policy Committee (the successor to the Working Group on Honours) as an appropriate solution. Thus, some seven years after the establishment of the Meritorious Service Cross, the letters patent were amended on June 6, 1991.
Recipients of the MSC and MSM are selected by either the Civil or the Military Advisory Committee of the Meritorious Service Decorations. The Canadian Forces Decoration Advisory Committee makes recommendations for award of the military Meritorious Service Decorations, while the Civil Advisory Committee discharges the same role in relation to the civil division. Both of the advisory committees are set up along the same lines as the Advisory Council of the Order of Canada.
Prototype Medal of the Order of Military Merit.
Prototype Meritorious Service Cross.
The MSC and MSM are awarded with certificates bearing the name of the recipient, the citation of the award, and the signature of the governor general.
The Meritorious Service Cross
Meritorious Service Cross (military) obverse.
Meritorious Service Cross (civil) obverse.
Criteria: Award of the cross's military division is open to all ranks of the Canadian Armed Forces — or allied forces — for the performance of a military deed or military activity in a highly professional manner or of a very high standard that brings benefit or great honour to the Canadian Armed Forces. The civil cross is awarded to Canadian citizens and Commonwealth and foreign citizens for the performance of a deed or activity in an outstandingly professional manner or of an uncommonly high standard that brings considerable benefit or great honour to Canada.
Insignia: Designed by Bruce Beatty, the insignia is identical for both the military and civilian divisions, consisting of a Greek silver cross with splayed and convexed ends surmounted with the Royal Crown. At the centre of the obverse is a maple leaf surrounded by a circle, and between the arms of the cross there is a laurel wreath. In the centre of the reverse is the Royal Cypher, and surrounded by a double circle, MERITORIOUS SERVICE MÉRITOIRE.
Meritorious Service Cross (military) reverse.
Meritorious Service Cross lapel pin.
Suspension: A straight graduated bar attached to the top of the crown.
Ribbon: For the military version the ribbon is 32 mm wide and blue in colour with a white stripe 6 mm wide centred on the outer third of each side of the ribbon. The civilian version is similar but with an additional 2 mm white stripe down the centre.
Bars: Subsequent awards of the cross are recognized with a silver bar with a maple leaf in the centre. A silver maple leaf is worn on the undress ribbon to represent a second award.
Naming: Capitals, initials, and surname are engraved on the reverse of the suspension bar, with the year of incident or event engraved on the uppermost arm on the reverse.
Lapel Badge: A small silver version of the insignia.
Postnominals: MSC.
Other: When established, the MSC could be awarded for deeds retroactive to June 11, 1984, whereas today the five-year rule applies so that a deed must have been undertaken at no more than five years prior to the submission of a nomination.
Number: 212 military division, 36 civil division.
The Meritorious Service Medal
Meritorious Service Medal (military) obverse.
Meritorious Service Medal (civil) obverse.
Criteria: Award of the military division of the Meritorious Service Medal is open to all ranks of the Canadian Armed Forces (or allied forces) for the performance of a military deed or military activity in a highly professional manner or of a very high standard that brings benefit or honour to the Canadian Armed Forces. The civil award is given to Canadian citizens and Commonwealth and foreign citizens for the performance of a deed or activity in a highly professional manner or of a very high standard that brings benefit or honour to Canada.
Insignia: Like the MSC, the medal was designed by Bruce Beatty and is identical for military and civilian recipients, consisting of a circular silver medal surmounted by the Royal Crown. The obverse bears the design of the Meritorious Service Cross and the reverse bears the Royal Cypher and MERITORIOUS SERVICE MÉRITOIRE.
Meritorious Service Medal (civil) reverse.
Meritorious Service Medal lapel pin.
Suspension: A small silver ring that passes through the top of the crown.
Ribbon: The military version has a blue ribbon 32 mm wide. Centred on the outer third of either side of the ribbon is a white stripe 6 mm wide with a 1 mm wide blue stripe in its centre. The civilian ribbon is similar but with the addition of a 2 mm white stripe down the centre of the ribbon.
Bars: Subsequent awards of the medal are recognized with a small silver bar with a maple leaf in the centre. A silver maple leaf is worn on the undress ribbon to represent a second award.
Naming: Capitals, initials, and surname are engraved around the rim, with the year of incident or event engraved on the reverse just below the crown.
Lapel Badge: A small silver version of the insignia.
Postnominals: MSM.
Other: When established, the MSM could be awarded for deeds retroactive to June 11, 1984, whereas today the five-year rule applies so that a deed must have been undertaken at no more than five years prior to the submission of a nomination.
Number: 787 military division, 292 civil division.
Bar for the Meritorious Service Cross/Meritorious Service Medal.
Insignia Manufacture: Both the MSC and MSM have been struck in sterling silver and also rhodium-plated metal. A variety of firms have produced the insignia: the Royal Canadian Mint, Rideau Ltée, Birks, and Pressed Metal Products. The last-named firm uses the mark PMP SS. Maker marks typically appear on the lower arm of the reverse. Beginning in 2015, the Royal Canadian Mint began producing the MSC and MSM in silver.
# 21
Mentions in Dispatches
The tradition of mentioning valuable services of a particular officer or man in dispatches to the Crown has a long history, dating back before the advent of gallantry or meritorious service medals intended for people other than senior officers. For a review of the early history of the Mention in Dispatches (MID) dating back to 1844, see chapter 5. During the First World War, more than five thousand Canadians were awarded MIDs, while the Second World War saw nearly ten thousand bestowed upon Canadians.
The MID was reinstated in Canada "for the purpose of according recognition to members of the Canadian Armed Forces on active service, and other individuals working with or in conjunction with the Canadian Armed Forces, for valiant conduct, devotion to duty or other distinguished services." While the MID had become a very junior award during the two world wars, it regained some of its prestige during the Korean War when so few bravery awards such as the Military Cross and the Military Medal were awarded. This situation prevails to this day: on most United Nations peacekeeping missions, relatively few, if any, bravery decorations are awarded, yet important service is rendered that warrants recognition and "mention."
The Canadian MID was established on November 1, 1990, as a national honour, though the actual letters patent were not signed until May 13, 1991. The insignia used during the Second World War and Korean War was retained and continues to be used. This form of recognition is more significant than many realize in that its eligibility is limited to operational service in the field, and in many ways it is akin to a junior Military Valour Decoration.
Recommendations for the MID are forwarded by the chief of the defence staff to the governor general. The CDS receives the recommendations from field commanders. Like other bravery and valour awards, the MID can be awarded posthumously.
Mention in Dispatches
Mention in Dispatches full-size oak leaf.
Mention in Dispatches certificate.
Criteria: For valiant conduct, devotion to duty, or other distinguished service in warlike situations in an active theatre of operations.
Insignia: The full-size insignia is a bronze oak leaf 28 mm long with prongs on the reverse for mounting. A 20 mm bronze oak leaf is worn on undress ribbons.
Wear: The insignia is worn on the designated ribbon. If the MID is awarded for service in a particular U.N. mission, the MID is worn on the ribbon of the U.N. medal awarded for that mission. If there is no designated ribbon, the insignia is worn on the left breast pocket. Prior to 1991, MIDs awarded without a designated ribbon were worn after medals or the undress ribbons.
Other: A recipient is also awarded an award certificate that summarizes the reason for the award (a citation); the certificate was originally signed by the governor general until September 2008. Since then it has been signed by the chief of the defence staff.
Number Awarded: 396 (311 of which were for Afghanistan).
# 22
The Sacrifice Medal
A number of countries bestow wound badges or medals upon those injured or killed in the line of duty. The general concept for the medal was aadvanced jointly by General Rick Hillier and then Minister of National Defence Gordon O'Connor, himself a retired brigadier-general. The creation of the Sacrifice Medal is one of the more controversial additions to the Canadian honours system. The overall concept for this medal has been questioned by a number of commentators.
Those who received wounds prior to October 7, 2001, will continue to be permitted to wear their wound stripes. However, those who received wounds after 2001 have been awarded the Sacrifice Medal and cease to be able to wear the wound stripe. The medal was developed between 2006 and 2008 and went through a number of variations. The proposed names ranged from the "Crimson Maple" to the "Medal of Sacrifice." At the time of publication plans were under way to announce the establishment of a Civilian Division of the Sacrifice Medal for bestowal on members of the protective services who are injured/killed in the line of duty.
Wound Stripe.
Within the Canadian Order of Precedence, the Sacrifice Medal is given precedence over all war and operational service medals and is worn immediately after decorations.
The medal was approved by Her Majesty the Queen on April 17, 2008, at Windsor Castle and was formally announced on August 29, 2008, with the first investiture taking place at Rideau Hall on November 12, 2009. The delay of more than a year between the announcement and the first presentation was the result of issues resulting from ineligibility of accidental deaths. At the direction of the minister of national defence, the chief of the defence staff conducted a review that resulted in broadening the original criteria to cover all service-related deaths and not just those that were the direct result of hostile action. These changes resulted in greater congruity between the posthumous aspect of the Sacrifice Medal and the criteria for the Memorial Cross, the Memorial Scroll, and the Memorial Bar.
The initial design concept for the medal and ribbon were developed by Major Carl Gauthier of the Department of National Defence. The final design and concept were developed by Cathy Bursey-Sabourin, Fraser Herald of the Canadian Heraldic Authority. The medal is struck in .999 silver by the Royal Canadian Mint.
The Sacrifice Medal replaces the wound stripe worn by members of the Canadian Armed Forces that dates back to the First World War and the Canadian Expeditionary Force. The wound stripe, originally named the wound strip, was authorized under a Canadian Army General Order of September 1, 1916, and the award of the wound stripe was initially made retroactive to August 4, 1914, the day Britain entered the First World War. The regulations were set out as follows:
> Strips of Russian braid, No. 1, two inches in length sewn perpendicularly on the left sleeve of the jacket to mark each occasion on which wounded. In the case of officers, the lower end of the first strip of gold braid will be immediately above the upper point of the flap on the cuff. Warrant officers, non-commissioned officers, and men will wear the gold braid on the left sleeve, the lower edge of the braid to be three inches from the bottom of the sleeve. The additional strips of gold braid, marking each subsequent occasion on which wounded, will be placed on either side of the original one, at half-inch intervals. Gold braid will be obtained free on indent from the Canadian Ordnance Corps; the sewing on will be carried out regimentally without expense to the public.[1]
In 1943 the Dress Regulations discontinued the wearing of wound stripes earned during the First World War. The Canada Army Overseas Routine Orders of April 6, 1944, reintroduced the wound stripe — a 4.5 cm narrow gold braid worn on the left forearm 12 cm above the bottom of the sleeve. Wounds received in previous wars were recognized with a single red rayon stripe, regardless of how many wound stripes the soldier had received prior to the Second World War.[2] Wound stripes existed in two forms — a piece of gold braid or a gilt metal bar that could be detached from the sleeve.
Wound Stripe certificate.
With the unification of the Royal Canadian Navy, the Canadian Army, and the Royal Canadian Air Force, many longstanding traditions were abolished, including the wound stripe. In 1995 the wound stripe was reinstituted for wear. As in the past, it consisted of one narrow gold braid, which was presented along with a certificate.
The Sacrifice Medal
Sacrifice Medal obverse.
Sacrifice Medal reverse.
Origins: This medal replaces the Wound Stripe Badge and is largely based on the Purple Heart awarded to members of the U.S. military.
Criteria: Awarded to members of Her Majesty's Canadian Armed Forces who have died as a result of military service or have been wounded as a direct result of hostile action. The extensive criteria includes civilian employees of the Government of Canada who are deployed as part of a military mission under the authority of the Canadian Armed Forces; friendly fire; death from wounds; exposure to nuclear, biological, or chemical agents; torture or maltreatment while a prisoner of war; and mental disorders, including post-traumatic stress disorder. The wounds must require not less than seven days' treatment in hospital. Self-inflicted wounds or accidental wounds are not covered by this medal. The next of kin receive both the Sacrifice Medal, which is awarded to the fallen Canadian Forces member, and the Memorial Cross, which is presented as a lasting memento of the person lost in the service of Canada.
Insignia: A 36 mm circular silver medal, the obverse depicting Her Majesty the Queen of Canada facing right wearing a Canadian diadem composed alternately of maple leaves and snowflakes and circumscribed with the legend ELIZABETH II DEI GRATIA REGINA and CANADA separated by small natural maple leaves. The reverse of the medal is a representation of the statute named "Canada" that forms part of the National Vimy Memorial, while facing right overlooking the horizon the inscription SACRIFICE appears along the edge of the lower right half of the medal.
Suspender: A suspender in the form of a Royal Crown attached to a straight slotted bar.
Ribbon: A 32 mm wide watered ribbon consisting of a 10 mm black stripe in the middle flanked by an 11 mm red stripe on each side on which are centred 1 mm wide white stripes.
Box for Sacrifice Medal.
Bars: A silver bar 31.5 mm by 6 mm with raised edges with a single natural maple leaf in the centre.
Naming: Engraved capitals, number, abbreviated rank, initials, and surname.
Postnominals: None.
Other: The medal is awarded in a maroon cardboard box the lid of which is stamped with the Royal Arms of Canada in gold. Sacrifice Medals bestowed at the inaugural investiture were presented in a special wooden box lined with black velvet, the lid of which is engraved with the Royal Arms of Canada. The chief of the defence staff has directed that this medal may only be presented by a flag or general officer, with delegation to a colonel or captain(N) only under exceptional circumstances.
Number: 813 plus 21 bars.
# 23
War, Operational, and Service Medals
War service medals are perhaps the most familiar awards to the general public. Many people can recall a father or a grandfather polishing his medals, and in most cases these were war medals. With the increase in the number of peacekeeping missions and Canada's involvement in military operations conducted by international organizations, however, the scope of contemporary war service medals has been broadened to include operations service medals and service medals. Chapter 6 provides an overview of the various imperial war medals that were awarded to residents of Canada who served in conflicts from the War of 1812 to the Korean War.
The first medal struck to recognize a conflict that involved the Canadian mili-tary was the North West Canada Medal, which was awarded to Canadian and British soldiers who served in suppressing the North West Rebellion. Another war medal unique to Canada was the Canada General Service Medal (1866–70), which was established in 1899 to reward those who had served in defending Canada during the Fenian Raids of 1866 and 1870 and also those who served in suppressing the Red River Rebellion in 1870.
While the Second World War yielded the Canadian Volunteer Service Medal and the Korean War a special Korea Medal, both created at the initiative of the Canadian government, these were still used within the broader imperial honours system in Canada. Although the British government had given Canada permission to use the British Naval General Service Medal and General Service Medal, this offer was never accepted. Nevertheless, a small number of Canadians serving on exchange with British and Australian forces received these medals.
The following chapter outlines the war, operational, and service medals that have been created since 1967. Earlier honours such as the aforementioned North West Canada Medal, Canada General Service Medal, Newfoundland Voluntary Service Medal, and Korea Medal are included in the first part of this book, since they were all integrated into the broader imperial honours system.
It was not until the 1990s that the Canadian government began instituting new medals on a regular basis. Prior to 1990, aside from the Canadian Forces' Decoration and United Nations Service Medals, there were no Canadian medals for operational service such as peacekeeping or for service with NATO or other international organizations.
Attempts were made throughout the late 1950s and early 1960s to establish a modern General Service Medal. The Canadian Armed Forces were involved in the International Commission for Supervision and Control in Indo-China, and there was a desire among some at the Department of National Defence to recognize their service with a Canadian medal. Regulations were created and designs were drawn up, always to be thwarted by the minister of national defence or a senior elected official. The most detailed of these proposals emerged in June 1957. It called for the creation of a circular silver medal with the Queen's portrait on the obverse and on the reverse an anchor, crossed swords, and an eagle superimposed on a maple leaf and circumscribed by GENERAL SERVICE GÉNÉRALE. This proposal served as the prototype for the Special Service Medal that was instituted in 1984 but not actually awarded until the early 1990s.
There are a variety of common elements among all the medals, be it an effigy of the Queen or the use of the Royal Crown, the Royal Cypher, and a variety of maple leaf motifs. The symbolic theme of including the effigy and cypher of the Sovereign along with other national symbols can be traced back to the first standardized war medal issued in 1812 when the Waterloo Medal was created.
The South-West Asia Medal was the first modern honour created that displays allegorical figures on the reverse of medals as opposed to the name of the medal or service. Similarly, the General Campaign Star, General Service Medal, and Operational Service Medal return Canada to more traditional medal designs.
As with all war, operational, and service medals, there are no postnominals associated with any of these honours.
The Canadian Volunteer Service Medal for Korea
Canadian Volunteer Service Medal for Korea obverse.
Origins: Some members of the Royal Canadian Navy, the Canadian Army, and the Royal Canadian Air Force who served in Korea did not arrive in the operational area until after the qualification period for the Queen's Korea War Medal concluded with the armistice on July 27, 1953. This meant that those who served prior to the armistice received both the Queen's Korea War Medal and the United Nations Service Medal — Korea, while those who served after July 1953 only received the U.N. Service Medal — Korea. This resulted in complaints from various veterans' groups about a lack of recognition, and a series of issues became conflated.
Although the Queen's Korea Medal was uniquely Canadian — bearing the text CANADA on the obverse and being created by the Crown in Right of Canada — some felt it was a British award and that those who served in Korea had never been recognized by the Canadian government. Added to this was the perception that while those who volunteered for service in the Second World War were awarded the Canadian Volunteer Service Medal, those who served in Korea received no recognition for their voluntary service.
During the Second World War not everyone was a volunteer and conscription was imposed in the later part of the war, so it was necessary to create some form of official recognition to acknowledge the voluntary nature of the service given by many. This was not the case in the Korea War, given that everyone who served was a volunteer. Despite the fact that Canada already had its own Korea War Medal and the reality that everyone who served in Korea volunteered, political pressure was brought to bear and this medal was created in 1991. In retrospect, the most seamless option would have been to extend the period of eligibility for the Queen's Korea War Medal.
Criteria: Those eligible are former members of the Canadian Armed Forces for service during all or part of the period beginning on June 27, 1950, and ending on July 27, 1954, in Korea, Japan, Okinawa, or in Korean waters, either on strength of an army unit of formation in Korea for at least one day, on active service for at least twenty-eight days on board a ship, or flew one sortie over Korea or Korean waters in the Yellow Sea or Sea of Japan, or accumulated at least twenty-eight days' service in the aforementioned areas.
Canadian Volunteer Service Medal for Korea reverse.
Insignia: A circular silver-coloured medal 36 mm in diameter made of a copper-and-zinc alloy and plated in rhodium. The obverse bears a crowned effigy of Her Majesty circumscribed by ELIZABETH II DEI GRATIA REGINA with CANADA at the base of the medal. The reverse of the medal bears the inscription KOREA VOLUNTEER 1950–1954 VOLONTAIRE CORÉE contained within a laurel wreath with a maple leaf centred at the bottom. The medal was designed by Bruce Beatty.
Suspender: A footed suspender mounted directly onto the top edge of the medal.
Ribbon: Yellow 32 mm wide bordered on either edge with a stripe 7 mm wide in United Nations blue. Down the centre there is a red stripe 6 mm wide on which is centred a 2 mm white stripe.
Bars: None.
Naming: None.
Number: 18,611.
Other: The medal was issued in a white rectangular cardboard box.
The Gulf and Kuwait Medal
Gulf and Kuwait Medal obverse.
Origins: In August 1990, Iraq invaded its neighbour Kuwait, an act that was immediately met with international indignation and outrage. Within a relatively short period of time a coalition was formed — led by the United States and aided by Great Britain, Saudi Arabia, Canada, Italy, and a variety of other partners — that sought to liberate Kuwait. Close to five thousand members of the Canadian Armed Forces were dispatched to the Persian Gulf. The majority were based in Bahrain or on board HMCS Athabaskan, Terra Nova, Huron, and Protecteur in the Persian Gulf. Canadian CF-18s also flew sorties over Iraq.
Criteria: Awarded to members of the Canadian Armed Forces (or those connected with it) who served a minimum of thirty days of cumulative service between August 2, 1990, and June 27, 1991, in Bahrain, Qatar, Kuwait, Iraq, or any country in the immediate vicinity in which members of the coalition forces were based. The medal and bar was awarded to those who served in the theatre of operations between January 16 and March 3, 1991.
Insignia: A circular silver-coloured medal 36 mm in diameter made of a copper-and-zinc alloy and plated with rhodium. The obverse bears a crowned effigy of Her Majesty circumscribed by ELIZABETH II DEI GRATIA REGINA with CANADA at the base of the medal. The reverse of the medal bears the inscription THE GULF AND KUWAIT 1990–1991 LE GOLFE ET KUWAIT contained within a laurel wreath with a maple leaf centred at the bottom. The medal was designed by Bruce Beatty. Interestingly, the French inscription contains a mistake in that the name KOWEÏT is spelled as in English.
Suspender: A footed suspender mounted directly onto the top edge of the medal.
Ribbon: The ribbon is 32 mm wide. A central stripe 8 mm wide and sand brown is flanked by a 5 mm navy blue stripe, a 2 mm scarlet stripe (to represent the army), and a 5 mm light blue stripe (to represent the air force).
Gulf and Kuwait Medal reverse.
Bars: A plain silver bar with a maple leaf in the centre.
Naming: None.
Number: 4,450 medals (3,198 with bar).
Other: Originally, the ribbon specified for this medal was to be the same used on the British Gulf War Medal. The manufacturer of the ribbon informed the Canadian government of this coincidence, and the design was slightly altered by reversing the two shades of blue. The medal was presented in a white rectangular cardboard box.
The Somalia Medal
Somalia Medal obverse.
Origins: In January 1991 the dictator of Somalia, Siad Barre, was overthrown and the country descended into chaos. As the situation deteriorated and the international community called for a ceasefire, the United Nations sent in a fifty-person observer team (UNOSOM) to monitor the situation. When the situation deteriorated further and starvation became widespread, the UNOSOM mission disintegrated. The U.N. Security Council subsequently adopted Resolution 794, which authorized "the use of all necessary means to establish as soon as possible a secure environment for humanitarian relief operations in Somalia." Operation Restore Hope and the United Task Force were authorized, and the Canadian Airborne Regiment was dispatched to Somalia. What transpired there resulted in the death of one Somali youth and the eventual disbandment of the Canadian Airborne Regiment. A small group of Canadian soldiers were implicated in the death of the Somali youth, and scandal erupted.
Initially, the Canadian government did not intend to issue a medal in light of the "Somalia Incident." As it became apparent that the failings in the theatre were more a reflection of significant structural problems within the leadership of the Canadian Armed Forces, the focus shifted away from the actual soldiers involved in the murder and onto the senior officers.
The award was instituted to recognize members of the Canadian Armed Forces who were involved in the Somalia Joint Task Force's Operation Deliverance. The mission's primary goal was to ensure peace and order in the capital city and other major centres, as well as to aid in the widespread distribution of food to the famine- and war-ravaged country.
The medal was sanctioned, but unlike previous Canadian campaign and service medals, the regulations required "honourable" service. Thus, those directly implicated in the killing of the Somali teenager were prohibited from receiving the medal.
Criteria: Awarded to members of the Canadian Armed Forces for a minimum of ninety days of honourable service (or sixty days if ordered out) in the Somalia Democratic Republic, the Republic of Kenya, or within two hundred nautical miles of the coast of Somalia between November 16, 1992, and June 30, 1993.
Insignia: A circular gold-coloured medal 36 mm in diameter. The obverse bears three maple leaves in fess and overlapping above which is inscribed CANADA and below which appear two sprigs of laurel leaves. The reverse depicts the Royal Cypher and Royal Crown circumscribed by SOMALIA SOMALIE 1992–93. The medal was designed by Bruce Beatty.
Suspender: A small, undetailed claw protrudes from the top of the medal through which a ring and the ribbon pass.
Somalia Medal reverse.
Ribbon: A U.N. blue ribbon with a central 22 mm white stripe is a series of five stripes, each of which are 2.5 mm wide, sand brown, navy blue, scarlet, light blue, and sand brown.
Bars: None.
Naming: None.
Number: 1,425.
Other: Early diagrams of the medal did not include CANADA, and the medal was originally supposed to have a straight suspender bar. The straight suspender bar was changed to a ring due to cost. The medal was issued in a white rectangular cardboard box.
Rotation Bars
Silver single-maple-leaf rotation bar.
Silver five-maple-leaf rotation bar.
Gold single-maple-leaf rotation bar.
Gold five-maple-leaf rotation bar.
The following group of overseas service medals — South-West Asia Service Medal, General Campaign Star, General Service Medal, and Operational Service Medal — include a new method of recognizing long periods of service in theatre through a system of rotational bars. Until recently our campaign and service medals were awarded to recognize participation in a campaign or operation regardless of its duration or the number of times a person deployed to it. Consequently, there was no mechanism to recognize the amount or number of times a person deployed to a specific mission
In June 1979 the United Nations created tour numerals for its medals, a system that NATO later emulated. Canadians soldiers earned these numerals over the years, creating an expectation that not only participation in an operation should be recognized but also its duration. The question did not arise for Canadian medals until 2003–04 because since the Korean War and the later creation of the U.N. and NATO numerals, no campaign recognized by a Canadian medal was long enough to justify such recognition.
With the Afghanistan campaign stretching to years and many Canadian troops redeploying for second and third rotations, the question of tour numerals came to the fore. When it was stated that there was no such mechanism for Canadian medals, much dissatisfaction ensued. After due consideration and a complete review of overseas service recognition, and keeping in mind the general aim of honours to not only recognize but also "encourage," which is vital in a military context, the regulations for the aforementioned medals were amended in 2010 to introduce a series of rotational bars to recognize extended periods of service in the mission.
The system of bars was seen as a more elegant and sensible way — given our method of mounting medals — than simply placing numerals on the medal ribbon following the U.N. and NATO practice. Rotation bars are awarded for each six months of additional accumulated service after the initial period of service required to earn the South-West Asia Service Medal, General Campaign Star, General Service Medal, or Operational Service Medal in question.
Insignia: There are four types of rotation bars. Silver rotation bars are awarded with the South-West Asia Service Medal, General Service Medal, and Operational Service Medal, while gold rotation bars are awarded with the General Campaign Star, the colour of the bar matching that of the medal in question. A bar bearing five maple leaves is awarded in lieu of five bars with a single maple leaf.
* • Silver Single-Maple-Leaf Bar: A silver bar 32 mm by 4 mm with a raised edge and a single natural maple leaf in the centre. Awarded for 180 days of additional service.
* • Silver Five-Maple-Leaf Bar: A silver bar 32 mm by 44 mm with a raised edge and five natural maple leaves. Awarded for 930 days of additional service.
* • Gold Single-Maple-Leaf Bar: A gold bar 32 mm by 4 mm with a raised edge and a single natural maple leaf in the centre. Awarded for 180 days of additional service.
* • Gold Five-Maple-Leaf Bar: A gold bar 32 mm by 44 mm with a raised edge and four natural maple leaves. Awarded for 930 days of additional service.
Rotational Bars
Period of Service | Medal or Bar(s)
---|---
30 days | South-West Asia Service Medal with Afghanistan Bar, General Campaign Star, General Service Medal or Operational Service Medal
210 days | One Single-Maple-Leaf Bar
390 days | Two Single-Maple-Leaf Bars
570 days | Three Single-Maple-Leaf Bars
750 days | Four Single-Maple-Leaf Bars
930 days | One Five-Maple-Leaf Bar
1,110 days | One Single-Maple-Leaf Bar and one Five-Maple-Leaf Bar
The award of rotation bars is denoted by the wear of maple leaf devices on the undress ribbon: silver for one bar, gold for two, red for three, and a combination for subsequent numbers such as a red and a silver for four, et cetera.
The South-West Asia Service Medal
South-West Asia Service Medal obverse.
Origins: After the terrorist attacks on New York City and Washington, D.C., on September 11, 2001, the international community rallied to root out the terrorist group Al Qaeda, which was principally located in Afghanistan. The United States, Canada, the United Kingdom, and other partners subsequently invaded Afghanistan and overthrew the pro–Al Qaeda Taliban government. The South-West Asia Service Medal was awarded to those who were involved in those operations.
Criteria: Awarded to members of the Canadian Armed Forces who were posted, attached to, or working with a Canadian or allied unit or a formation deployed outside Canada that was participating in or providing direct support on a full-time basis to operations conducted in Southwest Asia for a period of at least ninety days after September 11, 2001. Those who served thirty days inside the theatre of operations in Southwest Asia itself received a bar bearing the text AFGHANISTAN. Rotation bars are awarded for each additional 180-day period of service within theatre.
Insignia: A circular nickel-plated medal 36 mm in diameter. The obverse bears the effigy of Her Majesty the Queen wearing the King George IV State Diadem circumscribed with ELIZABETH II DEI GRATIA REGINA. At the base is CANADA flanked by maple leaves. On the reverse is a hydra, a multi-headed creature from Greek mythology, transfixed by a Canadian sword and circumscribed by the Latin phrase ADVERSUS MALUM PUGNAMUS ("We are fighting evil"). This concept was designed by Captain Carl Gauthier, while the formal artwork was created by Cathy Bursey-Sabourin, Fraser Herald at the Canadian Heraldic Authority.
Suspender: A straight suspender connects to the medal by a cluster of proportionate olive leaves.
South-West Asia Service Medal reverse.
Afghanistan Bar.
Ribbon: A white ribbon 32 mm wide flanked on each side by 4 mm of sand brown, 2 mm of red, and 4 mm of black at the outermost edge.The black represents the shock of the terrorist attacks of September 11 and the subsequent mourning for the victims. Sand brown represents the theatre of operations, white represents peace, and red represents the blood spilled on September 11.
Bars: Several types of bars are awarded with this medal:
* • AFGHANISTAN was awarded to individuals who served for a period of thirty days while deployed in the theatre of operations after September 11, 2001. In 2009, following a review of the recognition given to those who had served on multiple rotations, a series of rotational bars were introduced to recognize extended periods of service in theatre. The AFGHANISTAN Bar continues to be worn in addition to the rotational bar(s).
* • Silver Maple Leaf Rotation Bars were awarded for additional periods of service.
Naming: None.
Other: There are several issues of this medal. The first were manufactured by Carat Médailles and have a highly plated appearance. Later issues made by Bond Boyd Ltd. were struck in cupronickel using a new obverse die and have a much higher degree of detail and fineness. The first issues of this medal were bestowed in a red rectangular cardboard box, while all later issues were presented in a burgundy box embossed with the Royal Arms of Canada on the lid.
Number: 12,736 (12,409 with bar).
General Campaign Star
The practice of issuing a campaign star for service in a war zone or warlike situation can be traced back to the Honourable East India Company, which issued a campaign star for its various engagements in Gwalior, Maharajpoor, and Punnir in 1844. This tradition was continued with the Kabul to Kandahar Star that was awarded to British and Indian soldiers who served in the 480-kilometre relief march from Kabul to Kandahar led by Lord Roberts in 1880, and the 1914 and 1914–15 Stars issued to those who served in action during the early days of the First World War. A total of nine theatre-specific campaign stars were awarded during the Second World War. The main requirement to receive these various stars was a minimum period of service in a defined theatre of war. The practice of issuing a star to recognize war or warlike service is a tradition that continues in other Commonwealth countries such as India and Kenya.
In this tradition the General Campaign Star was created in 2004 to recognize services rendered in a warlike situation. At that time the Canadian honours system did not adequately recognize such military service, the Special Service Medal never having been intended for rewarding service in a combat zone. Initially, the General Campaign Star was issued with one of two bars: ALLIED FORCE and ISAF+FIAS. A person who served in both Operation Allied Force in the Balkans in 1999 and as a member of the International Security Assistance Force in Afghanistan from 2004 would have initially been awarded a single star and worn both bars on the ribbon. This system was amended in 2010 when it was decided to issue the General Campaign Star with theatre-specific ribbons, leaving room on these new ribbons for the new rotation bars. In addition to creating a new ribbon for ALLIED FORCE and keeping the original ribbon for SOUTH-WEST ASIA (in replacement of the more restrictive ISAF+FIAS bar), a more general ribbon called EXPEDITION was created to cover smaller operations in the presence of an armed enemy. Additional ribbons could be created for future major campaigns.
Insignia: A four-pointed gold star 44 mm wide, representing the four cardinal points of a compass. The obverse contains a wreath of maple leaves with the Royal Crown at the top. The blades and hilts of two superimposed crossed swords form four additional points to the star and over this are superimposed an anchor and a flying eagle. The Royal Cypher ensigned by the Royal Crown, a plain space for engraving, and three maple leaves on one stem are within a raised circle on the reverse. This concept was designed by Captain Carl Gauthier, while the formal artwork was created by Cathy Bursey-Sabourin, Fraser Herald of the Canadian Heraldic Authority.
Suspender: A small gold ring passes through a ball at the tip of the star.
Bars: Gold Maple Leaf Rotation Bars for additional periods of service. Prior to 2010, this medal was worn with the ALLIED FORCE or ISAF+FIAS bar.
Naming: Engraved capitals, number, abbreviated rank, initials, and surname on the reverse.
Other: Initially, these medals were struck by Bond Boyd Ltd. and then Carat Médailles. However, later issues have been struck by the Royal Canadian Mint. The medal is issued in a burgundy rectangular box whose lid is impressed in gold with the Royal Arms of Canada.
General Campaign Star — Allied Force
General Campaign Star — Allied Force obverse.
Gold Allied Force Bar (obsolete).
Origins: In 1999, NATO established Operation Allied Force, which undertook combat operations over Kosovo and other adjacent countries, notably the Federal Republic of Yugoslavia, Albania, and the Former Yugoslav Republic of Macedonia, as well as the Adriatic and Ionian Seas. Kosovo had previously been an autonomous province within the Federal Republic of Yugoslavia, which began disintegrating in the 1990s. In 1990, Kosovo declared itself a republic within Yugoslavia and then in 1992 declared independence, though it was only recognized by Albania. Tensions between ethnic Albanians in Kosovo (Kosovars) and the Serbian population resulted in the outbreak of civil unrest in 1998, with thousands of Kosovars being displaced. This was followed by the outbreak of the Kosovo War in 1999. Operation Allied Force sought to force the withdrawal of Serbian troops by bombing targets within the Federal Republic of Yugoslavia.
Criteria: Awarded to Canadian Armed Forces pilots and AWACS aircrew who flew at least five sorties over enemy territory during Operation Allied Force between March 24 and June 10, 1999.
Ribbon: Light blue 32 mm wide edged on each side by 2 mm of white and 8 mm of red.
Bars: None. Prior to 2009, this medal was worn with the ribbon now called SOUTH-WEST ASIA distinguished with the ALLIED FORCE bar now obsolete.
Number: 92.
Other: As Canadian honours policy precludes dual recognition, some recipients must exchange their previously awarded NATO Medals for the General Campaign Star.
General Campaign Star — South-West Asia
Gold NATO ISAF Bar (obsolete).
Origins: In 2001 the United Nations established a NATO-led security mission that became known as the International Security Assistance Force (ISAF). Initially, the mission was charged with securing Kabul and freeing it from Taliban and Al Qaeda control to allow for the establishment of the Afghan Transitional Administration. In 2003 the mission was expanded by the United Nations to include the entirety of Afghanistan in an effort to bring peace, stability, and the establishment of civil society in the war-torn country.
Criteria: At least thirty days of cumulative service in the Canadian Armed Forces as part of ISAF between April 24, 2003, and July 31, 2009, or in Afghanistan and the related naval theatre since August 1, 2009. Bars are awarded for subsequent periods of 180 days of service.
General Campaign Star South-West Asia obverse.
General Campaign Star — South-West Asia reverse.
Ribbon: Canadian Forces green 32 mm wide edged on each side by 2 mm of white and 8 mm of red.
Bars: Gold Maple Leaf Rotation Bars for additional periods of service. Prior to 2010, this medal was worn with the ISAF+FIAS bar, which is now obsolete.
Number: 32,343.
General Campaign Star — Expedition
General Campaign Star — Expedition obverse.
Origins: This medal is intended to recognize combat service in missions to which Canada makes a small contribution of personnel. To date two missions have been recognized with this medal:
* • Military service within Iraq any time after January 20, 2003. Canadian personnel on exchange with the American forces in Iraq are eligible for this medal, though those who serve with the United Nations Assistance Mission for Iraq (UNAMI) or who are on exchange with British forces are not eligible since their services are recognized by either the UNAMI U.N. Medal or the British Iraq Medal for Operation Telic.
* • Canadian personnel serving as part of Operation Impact, Canada's contribution to the fight against the Islamic State in Syria. Most recipients are air personnel flying over Iraq and Syria from bases in Kuwait.
Other missions can be added by the chief of the defence staff in consultation with the Armed Forced Council in the recommendation of the Canadian Forces Honours Policy Committee as the need arises to recognize service of Canadian Armed Forces members in combat-like missions that are not of significant enough scope or involvement to require a separate ribbon to the General Campaign Star.
Criteria: Awarded for thirty days of service to members of the Canadian Armed Forces who served in specific theatres of operation in approved locations outside Canada and were in the presence of an armed enemy, commencing January 1, 2003. Bars are awarded for subsequent periods of 180 days of service.
Ribbon: Light grey 32 mm wide edged on each side by 2 mm of white and 8 mm of red.
Bars: Gold Maple Leaf Rotation Bars for additional periods of service.
Number: 155.
General Service Medal
Origins: The idea to strike a new Canada General Service Medal dates back to the 1960s, though it was largely abandoned when the Special Service Medal was approved in 1984. This modern Canada General Service Medal is a companion to the General Campaign Star, the General Service Medal being reserved for operational support, while the General Campaign Star is for military service in theatre. Non-Canadian Armed Forces personnel deployed outside Canada either inside or outside the theatre of operations who provide direct support to the Canadian Armed Forces are eligible for this medal. This category includes members of the RCMP, Department of Foreign Affairs, and civilian contractors provided they are under the direct control of the Canadian Armed Forces. For military personnel, only support provided from outside the theatre is eligible since service in theatre is covered by the General Campaign Star. In all cases there needs to be a certain level of risk, threat, hardship, or operational intensity to justify the award of the medal.
As with the General Campaign Star, the initial General Service Medal was issued with the same green, white, and red ribbon along with bars specific for each operation: ALLIED FORCE or ISAF+FIAS. However, this system was changed in 2010 in favour of individual ribbons for each mission along with the new system of rotation bars that recognize extended periods of service in theatre.
Insignia: A silver-coloured circular medal bearing a crowned effigy of Her Majesty Queen Elizabeth II on the obverse circumscribed with ELIZABETH II DEI GRATIA REGINA and CANADA separated by small crosses pattée. These crosses represent military valour and merit and are of the shape used for the insignia of the Victoria Cross and the Order of Military Merit. On the reverse are two crossed swords, an anchor, and a flying eagle. The entire device is surrounded by two branches of maple leaves that form a wreath, and this is surmounted by a Royal Crown. The concept for this medal was designed by Major Carl Gauthier, while the formal artwork was created by Cathy Bursey-Sabourin, Fraser Herald of the Canadian Heraldic Authority.
Suspender: A straight bar suspender connects to the medal by a cluster of three overlapping maple leaves.
Naming: Engraved capitals, number, abbreviated rank, initials, and surname of military recipients; full given names and surname of civilian recipients.
Other: Each medal is issued in a burgundy rectangular box whose lid is impressed in gold with the Royal Arms of Canada.
General Service Medal — Allied Force
General Service Medal — Allied Force obverse.
Origins: This medal is the companion award to the General Campaign Star — Allied Force. It was established to recognize service in Operation Allied Force by members of the Canadian Armed Forces, which undertook aerial combat operations over Kosovo, the Federal Republic of Yugoslavia, Albania, the Former Yugoslav Republic of Macedonia, and the Adriatic and Ionian Seas in 1999. The medal was awarded to those who provided direct support to the mission and were deployed outside Canada but not necessarily in a theatre of operations. This medal may also be awarded to those who are not members of the Canadian Armed Forces who provided support either inside or outside the theatre of operations.
Criteria: At least thirty days of cumulative service as part of Operation Allied Force between March 24 and June 10, 1999. The medal was awarded to personnel who served in direct support of the operation in Aviano and/or Vicenza, Italy, or Skopje in the Former Yugoslav Republic of Macedonia.
Silver Allied Force Bar (obsolete).
Ribbon: Red 32 mm wide edged on each side by 2 mm of white and 5 mm of light blue.
Bars: None. Prior to 2010, this medal was worn with the ribbon now called SOUTH-WEST ASIA distinguished with the ALLIED FORCE bar now obsolete.
Number: 149.
General Service Medal — South-West Asia
General Service Medal — South-West Asia obverse.
Origins: This medal is the companion award to the General Campaign Star — South-West Asia. It was established to recognize service by members of the Canadian Armed Forces who served as part of ISAF in direct support of the mission but not necessarily in the theatre of operations. The medal may also be awarded to those who are not members of the Canadian Armed Forces who provided support either inside or outside the theatre of operations.
Silver NATO ISAF Bar (obsolete).
General Service Medal — South-West Asia reverse.
Criteria: At least thirty days of cumulative service as part of ISAF between April 24, 2003, and July 31, 2009, or those in direct support of Canadian military operations while deployed inside the theatre of operations commencing August 1, 2009. This includes service as part of Tactical Airlift Unit, Camp Mirage; ISAF liaison officer, Qatar; Logistics Support, Taranto, Italy; Interim Staging Team in Istanbul, Turkey; Intelligence Analyst, Coalition Intelligence Fusion Centre, Bahrain; Naval liaison officer, Bahrain; and Military Police Security Implementation Teams, Europe, North Africa, and others.
Ribbon: Red 32 mm wide edged on each side by 2 mm of white and 5 mm of Canadian Forces green.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service. Prior to 2010, this medal was worn with a silver ISAF+FIAS bar now obsolete.
Number: 5,774.
General Service Medal — Expedition
General Service Medal — Expedition obverse.
Origins: This medal is the companion award to the General Campaign Star — Expedition. It was established to recognize service in combat missions to which Canada makes a small contribution of personnel. As with the General Campaign Star — Expedition, other missions can be added as the need arises to recognize service of Canadian Armed Forces members and civilians who provide direct support to a specific mission.
Criteria: At least thirty days of cumulative service as part of a recognized mission. To date two missions have been recognized with this medal:
* • Military service in direct support of the U.S.-led Operation Iraqi Freedom, rendered from outside Iraq after January 20, 2003.
* • Canadian personnel serving in support of Operation Impact, Canada's contribution to the fight against the Islamic State in Syria, rendered from Kuwait support bases after September 4, 2014.
Ribbon: Red 32 mm wide edged on each side by 2 mm of white and 5 mm of light grey.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 7.
Operational Service Medal
The Operational Service Medal was created to recognize those members of the Canadian Armed Forces who serve or provide support to an overseas operation for which no U.N. or NATO medals have been created. The Operational Service Medal includes "theatre or task-specific ribbons" rather than a system of individual mission bars attached to a single medal. The eligibility and criteria for this medal is broader than that applied to the General Campaign Stars and General Service Medals. In addition to recognizing Canadian Armed Forces members, it can be used to recognize members of Canadian and allied police forces, public servants, and government contractors working as part of a sanctioned international mission or under dangerous circumstances overseas.
Insignia: A circular silver-coloured medal 36 mm in diameter struck in cupronickel, the obverse bearing an effigy of Her Majesty the Queen wearing the Canadian Diadem and circumscribed by the text ELIZABETH II DEI GRATIA REGINA and CANADA separated by two small natural maple leaves. The reverse displays a Royal Crown at the top along with sprigs of three natural maple leaves on each side, a Goode interrupted homosoline equal-area projection of the globe below, and a laurel branch crossed by an oak branch at the base.
Suspender: A non-swivelling claw-footed suspender with a straight bar.
Naming: Engraved capitals, number, rank abbreviation, initials, and surname of military recipients on the edge of the medal; for civilian recipients, full given name and surname on the edge of the medal.
Other: Medals are struck at the Royal Canadian Mint. The suspender is the same design used on the RCMP Long Service Medal.
Operational Service Medal — South-West Asia
Operational Service Medal, reverse, with Humanitas ribbon.
Operational Service Medal — South-West Asia obverse.
Origins: Awarded to civilian personnel who served in Afghanistan and adjacent areas and who were not recognized with the General Campaign Star — South-West Asia or the General Service Medal — South-West Asia. The medal was created primarily to recognize civilian and police service rendered under authority of an organization other than the Canadian Armed Forces. This includes Canadian diplomats working in Afghanistan.
Criteria: At least thirty days of cumulative service in Afghanistan, the Persian Gulf, the Gulf of Oman, the Gulf of Aden, the Red Sea, the Suez Canal, and parts of the Indian Ocean and Arabian Sea that are west of sixty-eight degrees east longitude and north of five degrees latitude, as well as the airspace above the aforementioned area, after October 7, 2001.
Ribbon: Sand brown 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 390.
Operational Service Medal — Sierra Leone
Operational Service Medal — Sierra Leone obverse.
Origins: In 1991, Sierra Leone descended into civil war when members of one of the country's political parties, the Revolutionary United Front (RUF), attempted to overthrow President Joseph Momoh. The RUF was supported by the government of neighbouring Liberia. The Lomé Peace Agreement was signed in 1999; however, the civil war lasted until 2002. A British-led International Military Advisory and Training Team was deployed to the country in 2000 to assist with the transition of the Republic of Sierra Leone Armed Forces into an accountable, self-sustaining, and affordable force to meet the defence needs of the country. Canadians who served as part of this mission prior to July 31, 2002, received the British Operational Service Medal with Sierra Leone ribbon.
Criteria: At least thirty days of cumulative service outside Canada providing direct support to operations conducted in Sierra Leone beginning July 31, 2002, if the service has not previously been recognized by a service medal other than the Canadian Peacekeeping Service Medal.
* • Participation in Operation Sculpture after July 31, 2002.
* • Canadian police officers working for the Special Court in Sierra Leone are also eligible for this medal.
Ribbon: Light green 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 122.
Operational Service Medal — Haiti
Operational Service Medal — Haiti obverse.
Origins: Awarded to Canadians who have served in Haiti or have provided direct support to operations conducted in Haiti provided the service has not been recognized with the award of a U.N. medal, though the recipient may have been recognized with the Canadian Peacekeeping Service Medal.
Criteria: At least thirty days of cumulative service beginning March 6, 2004. This includes those who have served as part of:
* • Operation Halo, the U.S.-led Multinational Interim Force, from March 6 to August 16, 2004.
* • Members of the medical evacuation team deployed to the Dominican Republic between March 6 and August 16, 2004.
Ribbon: Royal blue 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 926.
Operational Service Medal — Sudan
Operational Service Medal — Sudan obverse.
Origins: Sudan has been riven with armed conflict for most of the past fifty years, beginning shortly prior to independence being granted by Britain and Egypt in 1956. There have been three conflicts: First Civil War, 1955–72; Second Civil War, 1983–2005; and the most recent iteration known as the Sudan Internal Conflict, which commenced in 2011 as a result of South Sudan being established as an independent country. Canada has provided various degrees of support to enhance the peace and security of the region in co-operation with other international partners.
Criteria: At least thirty days of cumulative service as part of one of the following recognized missions:
* • Operation Augural in Sudan as part of the African Union's Darfur Integrated Task Force between September 15 and December 31, 2007.
* • Service of Canadian Armed Forces personnel deployed to the Canadian embassy in Khartoum as part of the National Support Element or National Communications Centre, which commenced supporting operations in Sudan on September 15, 2004. This includes personnel deployed to Ethiopia and Senegal who provide direct support to the mission.
* • Service of sworn Canadian police officers working as part of the United Nations Development Programme in South Sudan beginning September 17, 2009.
Ribbon: Dark green 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 125.
Operational Service Medal — Humanitas
Operational Service Medal — Humanitas obverse.
Origins: This honour replaces the Special Service Medal with HUMANITAS bar as of August 1, 2009. It is awarded for service in a humanitarian mission conducted outside Canada in response to a disaster or human conflict, including rescue, relief, and reconstruction.
Criteria: At least thirty days of cumulative service in an approved humanitarian mission on or after August1, 2009. To date three missions have been recognized:
* • Operation Hestia, the humanitarian mission in Haiti between January 12 and May 2, 2010. The mission was established after a devastating earthquake suffered by the country on January 12, 2010.
* • Operation Renaissance, the humanitarian mission in the Philippines that commenced on October 15, 2013, following a catastrophic typhoon suffered by the country on October 15, 2013.
* • Operation Renaissance 15-01, the humanitarian mission in Nepal, commencing on April 25, 2015, following a devastating earthquake suffered by the country on April 25, 2015.
Ribbon: White 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 1,198.
Operational Service Medal — Expedition
Operational Service Medal — Expedition obverse.
Origins: As with the General Campaign Star — Expedition, and the General Service Medal — Expedition, this award is intended to recognize those serving in operations to which Canada makes a small contribution, be it the Canadian Armed Forces, police, or civilian personnel.
Criteria: At least thirty days of cumulative service as part of a recognized operation. The list of recognized operations currently includes:
* • Sworn Canadian police officers deployed in Iraq providing assistance in the establishment and training of the new Iraqi police beginning January 20, 2003.
* • Members of the Defence Team deployed to the Middle East assisting the Office of the United States Security Coordinator (USSC) in Operation Proteus beginning May 3, 2005. Deployment of a Canadian Armed Forces team in Jerusalem to support the USSC providing the Palestinian Authority Security Forces with training, advice, and support, and to support the construction of security infrastructure for the Palestinian Authority.
* • Service as part of Operation Chabanel in the theatre of operations consisting of the waters off the west coast of equatorial Africa in the Gulf of Guinea from April 17 to May 18, 2006. Assistance offered to the RCMP in drug interdiction in the South Atlantic. HMCS Fredericton seized 22.5 tons of hashish.
* • Service with Operation Caribbe under the commander of the Joint Interagency Task Force South in the theatre of operations consisting of the Caribbean Sea/Gulf of Mexico and along the west coast of Central and South America, including airspace above this area, beginning November 1, 2006. An ongoing operation that interdicts the smuggling of illicit drugs, money, weapons, and people in the Caribbean.
* • Service with the United Nations International Independent Investigation Commission Mission in Beirut, Lebanon, from February 13, 2008, to February 28, 2009. This mission investigated the assassination of the former prime minister of Lebanon, Rafic Hariri, who was killed on February 14, 2005.
Ribbon: Light grey 32 mm wide edged on each side by 2.5 mm of white and 2.5 mm of red.
Bars: Silver Maple Leaf Rotation Bars for additional periods of service.
Number: 4,819.
The Special Service Medal
Special Service Medal prototype.
Special Service Medal cupronickel issue obverse.
Origins: This medal was originally instituted in 1984 after nearly three years of work. It received initial approval but was then set aside. The original medal was to be struck in cupronickel and have a straight suspender bar that would be separately attached, while the individual bars were to be riveted to the medal, similar to the way clasps are attached to the British General Service Medal. The original regulations indicated that each medal was to be named, and this remained in place until the regulations were changed in 1992. However, no medal was ever officially named. Recipients were also to receive special certificates of award with the medal, but none has been issued.
The medal was revived in 1992 around the same time that the Canadian Volunteer Service Medal for Korea and the Gulf and Kuwait Medals were being developed. The only difference between the regulations of 1984 and those issued in 1992 was that the 1984 regulations related more to the administration of the medal than to the medal itself. Each bar is authorized by an order-in-council, and as of 2013, seven bars have been authorized, though one — JUGOSLAVIJA — was subsequently revoked and never issued.
Consideration was given to bars for the FLQ Crisis of 1970 and Oka Crisis of 1990. However, no bars were created for these incidents, the prevailing view having long been that services rendered within Canada during peacetime emergencies is simply part of the normal domestic obligations of the Canadian Armed Forces ("Canadians helping Canadians"), being a civic obligation that is shared by all citizens in various capacities. The medal is awarded only to members of the Canadian Armed Forces and can be awarded posthumously.
Insignia: A circular silver medal 36 mm in diameter made of a copper-and-zinc alloy and plated in rhodium. The obverse bears a single maple leaf surrounded by a laurel wreath, while the reverse depicts the Royal Cypher and Royal Crown in the centre circumscribed with SPECIAL SERVICE SPÉCIAL. The medal was designed by Bruce Beatty.
Criteria: Awarded for special service in a specific theatre. The criteria vary according to the bar. The medal is never issued without a bar.
* • PAKISTAN 1989–90: The recipient must have given a minimum of ninety days of service with the Mine Awareness and Clearance Training Program in Pakistan during the period beginning March 15, 1989, and ending July 29, 1990, under the auspices of the United Nations. This bar was created in 1992 and can now be exchanged for the United Nations Special Service Medal.
* • PEACE — PAIX: The recipient must have served an aggregate of 180 days of honourable service in peacekeeping operations from the beginning of international peacekeeping operations in November 1947 through July 21, 2001. The service cannot have been recognized by any other award in, or accepted by, the Canadian system of honours. The list of missions for which this bar can be awarded consists of every U.N. peacekeeping mission and international mission in which Canada has been involved. This bar was awarded to those members of the Canadian Armed Forces who participated in peacekeeping missions yet did not meet the minimum requirement to qualify for the related U.N. service medal. One could not receive the U.N. medal and the PEACE bar for the same mission. This bar was created in 1992, and the criteria were subsequently revised in 1995. In 2001 this bar was superseded by the Canadian Peacekeeping Service Medal.
Special Service Medal cupronickel issue reverse.
* • ALERT: The recipient must have served an aggregate of 180 days of honourable service on the posted strength of Canadian Armed Forces Station Alert or have performed honourable service with a military force operationally deployed to or at CFS Alert, since it began its operations on September 1, 1958. This bar was first created in 1992, and the criteria were revised in 1995.
* • NATO+OTAN: The recipient must have served an aggregate of 180 days of honourable service within the North Atlantic Treaty Organization's area of responsibility between January 1, 1951, and October 19, 2004, defined as the territory of any of the parties in Europe or North America, the Algerian departments of France, or the territory of Turkey, or the islands under the jurisdiction of any of the parties in the North Atlantic north of the Tropic of Cancer. Service from October 20, 2004, must total 180 days of honourable service performed in approved locations or tasks outside Canada as part or in direct support of NATO operations or missions, provided the mission is not recognized by another medal. To date, eligibility is limited to:
Special Service Medal first issue obverse.
* * Service on board ships deployed under NATO Standing Naval Force Atlantic (Standing NATO Reaction Force Maritime Group 1).
* * Aircrews flying NATO surveillance missions outside Canadian airspace as part of an actual NATO mission outside the territorial waters and airspace of Canada.
* * CC NATO Airborne Early Warning Force (AWACS aircrews on NATO missions).
* * Service in support of Operation Active Endeavour (Canadian Operation Sirius) provided that the service was not already recognized by the award of the NATO Article 5 Medal for Active Endeavour (days of service multiplied by six to equate the NATO Medal criteria of thirty days).
* * Service with the Forward Logistics Site (FLS) in support of NATO operations and activities in Africa from January 1, 2008, provided service has not already been recognized by the award of the NATO Non-Article 5 Medal for service in Operations Allied Provider, Allied Protector, or Ocean Shield (days of service multiplied by six to equate the NATO Medal criteria of thirty days).
All staff, logistical, and administrative support duties as well as training are now considered normal military duty and are excluded from eligibility. A person must have been under NATO command for 180 days in order to qualify for this bar. The bar was first created in 1992, and the criteria were subsequently revised in 1995 and again in 2004.
Pakistan Bar.
Peace Bar.
Alert Bar.
NATO Bar.
* • HUMANITAS: This bar was established in 1993 and was superseded in 2009 by the Operational Service Medal — Humanitas. Prior to 2009, it was awarded to those who performed an aggregate of thirty days of honourable service outside Canada after June 11, 1984, in support of any humanitarian operation conducted in response to a disaster or human conflict, including rescue, relief, and reconstruction operations. The service cannot be acknowledged by the award of any other honour that is part of the Canadian honours system. This award includes the following missions and programs:
* * Ethiopia: Medical teams and relief supplies, February 1985.
* * Mexico: Earthquake relief, September 1985.
* * Colombia: Relief supplies after the mudslide, November 1985.
* * Ethiopia: Famine relief, June 1988.
* * Armenia: Provision of relief supplies, December 1988.
* * Jamaica: Relief supplies, September 1988.
* * Montserrat and Nevis: Relief supplies and medical assistance, October 1991.
* * Turkey and Iraq: Post–Gulf War medical assistance to Kurds, March 1991.
* * Ethiopia: Famine relief, August 1991.
* * Commonwealth of Independent States: Humanitarian assistance, January 1992.
* * Florida: Humanitarian relief after a hurricane, September 1992.
* * Bahamas: Humanitarian relief following a hurricane, October 1992.
* * Rwanda: Humanitarian relief to refugees, June 1994.
* * Somalia: Airlift of relief supplies, August 1992 to January 1993.
* * Zaire: Humanitarian relief, November 1996 to January 1997.
* * CARE Canada: Canadian Armed Forces members seconded to CARE where three months are served overseas in a humanitarian mission, 1998 to present.
* * Italy: Search equipment provision to aid Sarno, Italy, following mudslides, May-June 1998.
* * Canadian International Development Agency (CIDA): Department of National Defence personnel seconded to CIDA working on humanitarian operations, 1998 to present.
* * Central America: Emergency medical supplies following Hurricane Mitch, November-December 1998.
* * Turkey: Disaster Assistance Response Team (DART) providing earthquake relief, August-September 1998.
* * Albania: Canadian Armed Forces support to rehabilitate Rinas airfield near Tirana, March-September 2001.
* * Southern United States of America: Humanitarian relief following Hurricane Katrina, September-October 2005.
* * Pakistan: Humanitarian relief and DART following earthquake, October 2005 to January 2006.
* * Lebanon: Aiding the evacuation of Canadian citizens, July-September 2006.
* * Haiti: Humanitarian service following hurricanes, September 2008.
Humanitas Bar.
* • RANGER: Awarded for an aggregate of four years of honourable service as a Canadian Ranger performing the duties involved with providing a military presence in support of Canadian sovereignty. Eligible service includes reporting unusual activities, collecting local data of significance to support military operations, supplying local expertise or assistance and advice, acting as a guide or adviser in search-and-rescue activities, and completing a minimum of three Ranger Patrol Exercises within Canada or its territorial and contiguous waters since 1947. This bar was created in 1999.
* • EXPEDITION: Awarded for an aggregate of 180 days of honourable service on or after July 1, 2007, while deployed outside Canada on a temporary basis for the purpose of participating in or providing direct support on a full-time basis to approved operations. This bar was created in 2014.
* * Service in Comalapa, El Salvador, Curaçao, and the liaison officer in Key West, Florida, in direct support of the U.S.-led Joint Task Force — South mission, since July 1, 2007 (Operation Caribbe).
* * Service as a staff member at the Third Location Decompression Site, Cyprus, since August 1, 2007.
* * Service with the Integrated Undersea — Surveillance Towed Array Sensor System while deployed on U.S. Navy ships performing surveillance in the South China, East China, and Yellow Seas as well as part of the Sea of Japan, since December 1, 2007.
* * Service with the Casualty Support Team, Landstuhl Regional Medical Centre, Germany, from February 11, 2008, to March 29, 2014.
* * Service with the U.S. Navy Central, Naval Support Activity, Manamah, Bahrain, including but not limited to:
* ․ Combined Maritime Force (CMF) Headquarters, since April 21, 2010.
* ․ 5th Fleet Headquarters, since July 31, 2012.
* ․ Coalition Intelligence Fusion Centre, since May 21, 2014.
* ․ United States Naval Forces Central Command (NAVCENT) liaison officer, since May 21, 2014.
* ․ Service at the Strategic Lines of Communications Detachments in Germany, since October 1, 2010; in Cyprus, since October 21, 2010; and in Kuwait, since June 2, 2011.
* ․ Service with the U.S. forces in Kuwait in direct support of the Iraq transition to provide advice, to assist with transition from Department of Defence to State Department, and to assist with the retrograde of U.S. forces out of Iraq from February 1 to December 31, 2011.
* ․ Service with Task Force Jamaica within the political boundaries, territorial waters, and airspace of Jamaica in support of the Jamaica Defence Force to provide search-and-rescue and Medevac capabilities from August 12 to November 15, 2011 (Operation Jaguar).
* ․ Service with Air Task Force Mali in support of the French-led Operation Serval from the Istres–Le Tube Air Base in France from January 15 to April 3, 2013.
* ․ Service with United States Central Command (USCENTCOM) Forward — Jordan (CF-J), King Abdullah Special Operations Training Centre (KASOTC), Amman, Jordan, since May 23, 2013.
Expedition Bar.
Ranger Bar.
Suspender: A straight slotted bar mounted directly onto the top edge of the medal.
Ribbon: CF green ribbon 32 mm wide edged on either side with 5 mm of white and 5 mm of red. A small silver, gold, or red maple leaf is worn on the undress ribbon to indicate earning more than one bar.
Bars: Individual bars 32 mm long are sewn onto the ribbon. To date, seven bars have been approved but only six actually awarded as described above.
Naming: None.
Number:
* • PAKISTAN 1989–90: 50.
* • PEACE — PAIX: 2,321.
* • ALERT: 7,902.
* • NATO+OTAN: 64,670.
* • HUMANITAS: 1,120.
* • RANGER: 3,236.
* • EXPEDITION: 148.
Other: The original strikings of this medal are heavily plated and were undertaken by a number of private firms. Since 2011 all Special Service Medals have been struck in cupronickel by the Royal Canadian Mint and have a much higher degree of detail. The medal was originally issued in a white rectangular cardboard box with the name of the medal in black letters on the lid. The more recent strikes are issued in the standard burgundy presentation box bearing the Royal Arms of Canada in gold on the lid.
The Canadian Peacekeeping Service Medal
Peacekeeping Service Medal obverse.
Origins: The Canadian Peacekeeping Service Medal was established in part so that Canadian peacekeepers would receive an award from Canada besides the service medals they receive for a specific mission from the overseeing international organization (United Nations or NATO). The medal was established to acknowledge the award of the 1988 Nobel Peace Prize to all U.N. peacekeepers. In that same year the Department of National Defence considered creating a special medal for those who had served in peacekeeping operations, but it was decided at the time that no medal should be struck, since those who had participated in U.N. missions and completed the requisite period of service had already been awarded one of the U.N. service medals. To award two medals for one action is contrary to Canadian honours policy, and since this would have happened in the majority of cases, the plan was shelved.
In late 1996 a private member's bill was introduced into the House of Commons by Jack Fraser, OMM, MSC, CD. Parliament subsequently passed an act sanctioning the creation of this medal, making it the only medal in the Canadian honours system established through an act of Parliament.
The Canadian medal is more accessible to peacekeepers, given that it is awarded for thirty days of service as opposed to the ninety to 180 days required for most U.N. peacekeeping service medals. Those who had previously not qualified for a U.N. medal or the Special Service Medal with the PEACE bar thus became eligible for the new medal. Civilians can also be awarded the medal, though it is primarily awarded to members of the Canadian Armed Forces.
Criteria: A minimum of thirty days of cumulative peacekeeping service as a member of a force under the auspices of the United Nations or with another recognized international peacekeeping force.
Peacekeeping Service Medal reverse.
Insignia: A circular medal made of an oxidized silver-coloured alloy. The obverse depicts the three Canadian peacekeeper figures atop the Peacekeeping Monument in Ottawa, and this is circumscribed by PEACEKEEPING — SERVICE DE LA PAIX together with two maple leaves. The reverse bears the Royal Cypher and Royal Crown superimposed upon a maple leaf surrounded by two sprigs of laurel. CANADA appears at the base. The medal was designed by Bruce Beatty. The obverse design is taken from the Canadian Peacekeeping Memorial in Ottawa, which was designed by Jack Harman.
Suspender: A footed suspender bearing a maple leaf in the centre mounted directly onto the top edge of the medal.
Ribbon: United Nations blue 32 mm wide edged on each side by 4 mm stripes of white, red, and green.
Bars: None.
Naming: None.
Number: Approximately 75,521 to date. Approximately 130,000 personnel are eligible.
Other: The obverse of the medal was originally intended to be the side that depicts the maple leaf and Royal Cypher, as is the usual practice. Rather suddenly, after pressure from the Canadian Association of Veterans in United Nations Peacekeeping, this was changed so that the side depicting the peacekeeping monument became the obverse. The medal is issued in a white rectangular cardboard box with the name of the medal in black letters on the lid.
# 24
The United Nations Service Medals
The United Nations is perhaps the world's most widely known organization, its flag being recognized around the globe. Founded in 1945, the United Nations has served as a focal point for international diplomacy and peaceful co-operation among nations. The institution has gone to war on two occasions — the Korean War and the First Gulf War of 1990–91 — yet it is most widely known for its peacekeeping, observing, and monitoring missions.
Under the auspices of the United Nations, fifty-seven separate medals have been issued for these missions. In nearly all of these cases, it is only the ribbon and not the physical metal medal that changes from one mission to another, the United Nations preferring to spend its resources on the various missions and not on medals. In this sense the U.N. medals are similar to the campaign stars issued during the Second World War.
The design of the various types of U.N. medals is relatively simple. The symbols employed on each medal are very similar, with only minor changes in wording or the suspension style used. Since the Korean War, the ribbons issued have generally been representative of the national colours of the country or countries in which the particular mission took place; early on it was discovered that there was a finite number of ways that blue and white could be used to make different ribbons. Even today, with fifty-seven different ribbons issued, it is difficult to develop new ribbon designs that are not easily mistaken for those that have already been issued. Indeed, some ribbons are shared between missions, while some U.N. medals have up to five different bars denoting different segments of a certain extended mission (Haiti).
The focal point of the United Nations' emblem is a map of the globe centred on the North Pole and drawn using an azimuthal equidistant projection (one in which a straight line drawn from the centre to any point on the map will always represent the shortest distance between the two points) surrounded by olive branches. The pale blue used by the United Nations on its flag and in all U.N. medal ribbons — United Nations blue, less commonly known as Stettinius blue — was selected because no country used that colour in its national flag at that time. The U.N. emblem and colour scheme were designed by the United States State Department and were unveiled at the first U.N. Conference held in San Francisco in 1945.[1]
The first U.N. medal, now known as the United Nations Service Medal (Korea), was established in October 1950 for service in the Korean War. A total of 2,760,000 United Nations Korea Medals were issued, with eleven different languages being struck: English, Amharic, Dutch, French, Greek, Italian, Korean, Spanish, Tagalog, Thai, and Turkish.
Despite the alacrity with which the United Nations Korean Medal was devised and approved, it would not be until the mid-1960s that a standard medal was sanctioned for other U.N. missions. The standard-issue U.N. medal — the United Nations Service Medal — was authorized by the secretary general on February 16, 1966, and is currently on issue. As noted, the medal remains the same for each mission, while the ribbon differs, making this medal the most widely distributed multi-theatre award ever issued. There are three types of U.N. medals — only the standard type is currently on issue — and to date the United Nations has issued them for fifty-seven missions, of which Canadians have been active in forty-three. In addition to these mission-specific medals, there has also been a U.N. medal issued for service at U.N. Headquarters in New York City and there is a United Nations Special Service Medal for missions for which no mission-specific medal is created.
Although the U.N. medals are created by the United Nations, it is not until the Canadian government approves them for wear and integrates them into the Canadian Order of Precedence that Canadians are permitted to wear them. Only medals issued for those missions that Canada is officially involved in are approved. Once approved, the medals are considered "Canadian" awards.
This chapter will not recount the historical background behind each U.N. mission that has been recognized with a medal; such an undertaking has already been completed in books like In the Eye of the Storm: A History of Canadian Peacekeeping by Fred Gaffen and The Canadian Peacekeeper by Colonel John Gardam. Instead, the goal of this chapter is to provide a broad overview of the criteria and basic events surrounding each mission.
There is no master register of record of every U.N. medal issued to Canadians, and precise numbers of medals issued for each mission are therefore difficult to obtain. This ambiguity is due in part to the fact that the medals are issued in theatre and the awards are recorded on each individual's record. There are no postnominals associated with any of these honours.
United Nations Service Medal Korea
U.N. Service Medal Korea obverse.
U.N. Service Medal Korea English reverse.
Criteria: One day on active service under United Nations Command in Korea and in the adjacent areas, including Japan and Okinawa, between June 27, 1950, and July 27, 1954. The medal could also be awarded for an aggregate of thirty days of service spent on official visits of inspection to the qualifying area.
Insignia: The United Nations Korea Medal is bronze and 35 mm in diameter. On the obverse the U.N. emblem appears in bas-relief, while the reverse bears the legend FOR SERVICE IN DEFENCE OF THE PRINCIPLES OF THE CHARTER OF THE UNITED NATIONS — or in French, POUR LA DEFENSE DES PRINCIPES DE LA CHARTE DES NATIONS UNIES.
Suspender: A single-toe claw straight bar suspender and bar with KOREA or CORÉE in bas-relief.
Ribbon: A 35 mm wide ribbon made up of seventeen alternating equal stripes of white and U.N. blue.
Naming: Impressed capitals, number, initials, and surname. This is the only U.N. medal that was named when issued to Canadians.
United Nations Medals
There have been two issues of the United Nations Service Medals. The first was awarded solely for service in the United Nations Emergency Force (UNEF) from 1956 to 1967. The second and most common issue has been awarded for every other recognized U.N. mission. The standard U.N. Medal issue is as follows.
Insignia: A circular bronze medal 35 mm in diameter. On the obverse UN appears above the United Nations emblem. On the reverse IN THE SERVICE OF PEACE appears.
Suspender: A small ball mounted on top of the medal through which passes a mounting ring and the ribbon.
Bars: A number of bars have been issued, usually only on a temporary basis until a proper unique ribbon could be developed for a particular mission. For example, U.N. forces serving in the Congo were initially awarded a United Nations Service Medal with the same ribbon as the United Nations Truce Supervision Organization (UNTSO) issue, but with a bar with CONGO embossed on it to set it apart. A distinctive ribbon for Opération des Nations Unies au Congo (ONUC) was developed a few years later. The United Nations Special Service Medal and the U.N. medal for the various U.N. missions in Haiti are accompanied by approved bars.
Numerals: A small silver numeral is worn on the ribbon for additional tours. The United Nations approved numerals in 1979.
Other: The medals have been manufactured by upward of twenty-five different firms around the world, and the quality of the metal sometimes varies. U.N. medals are not typically issued with a certificate, though unofficial "in-theatre" certificates are sometimes presented.
United Nations Emergency Force (UNEF), 1956–67
UNEF Medal.
UNEF Medal reverse (standard to all U.N. medals).
Origins: After Egypt nationalized the Suez Canal in 1956, British, French, and Israeli troops combined to launch an assault. In November of that year the United Nations Emergency Force (UNEF) headed by Canadian General E.L.M. Burns was organized and dispatched to occupy the Canal Zone, enforce the ceasefire that had been declared, supervise the withdrawal of foreign troops from Egypt, and act as a buffer between Egyptian and Israeli forces. At Egypt's request the mission ended in June 1967.
Criteria: Ninety days of service between November 7, 1956, and May 19, 1967.
Insignia: Identical to the standard U.N. Medal. However, UNEF appears on the obverse above the U.N. emblem.
Ribbon: A 35 mm wide ribbon sand brown in colour with a 9 mm central stripe of U.N. blue flanked on each side by 1 mm stripes of green, sand brown, and dark blue.
Number: 9,963.
United Nations Truce Supervision Organization in Palestine (UNTSO), 1948 to Present
UNTSO Medal.
UNGOMAP Medal.
Origins: The United Nations' first peacekeeping initiative, the United Nations Truce Supervision Organization (UNTSO), was set up in June 1948 to oversee a ceasefire between Israel, Egypt, Jordan, Lebanon, and Syria. Subsequently, it has carried out various tasks in the region, including the containment of hostilities and the enforcement of truces after the Suez Crisis of 1956 and the Six Day War of 1967. UNTSO continues to support and co-operate with U.N. missions — United Nations Disengagement Observer Force (UNDOF) and United Nations Interim Force in Lebanon (UNIFIL) — in the region and has provided observers to help set up several other missions: United Nations Good Offices Mission in Afghanistan and Pakistan (UNGOMAP), United Nations Iran-Iraq Military Observer Group (UNIIMOG), United Nations Iraq-Kuwait Observation Mission (UNIKOM), and United Nations Protection Force (UNPROFOR).
Criteria: Ninety days of consecutive service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue superimposed with 1 mm white stripes, inset 6 mm from each edge.
Number: 920.
United Nations Observer Group in Lebanon (UNOGIL), 1948–58
Origins: At the request of Lebanese president Camille Chamoun, the Security Council formed this observer group in June 1958 to ensure that unauthorized personnel, arms, and other matériel did not enter the country during its civil war. The mission, primarily staffed by U.S. troops, ended in December of the same year.
Criteria: Thirty days of consecutive service between June 11 and December 9, 1958.
Insignia: Standard U.N. Medal.
Ribbon: Same as for UNTSO.
Number: 77.
United Nations Military Observer Group in India and Pakistan (UNMOGIP), 1948–79
Origins: After the partition of India and Pakistan and the granting of independence to the two states, violence broke out between them over the princely state of Jammu and Kashmir. UNMOGIP was established in January 1949 to oversee the cessation of hostilities and observe the ceasefire line drawn through the region, which was divided between the two countries.
Criteria: Service of 180 consecutive days between January 20, 1948, and January 20, 1979.
UNMOGIP Medal.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon shaded green and edged by 0.5 mm of white and 4 mm of U.N. blue.
Number: 501.
Opération des Nations Unies au Congo (ONUC) — United Nations Operation in the Congo, 1960–64
ONUC Medal.
Origins: In the summer of 1960, the Congo gained its independence from Belgium, an event marked by the mutiny of the Congolese army and an attempt by Katanga Province to secede. Although the Congo was no longer a colony, the Belgian government in Brussels dispatched troops to the fledgling nation. At Prime Minister Lumumba's request the U.N. Security Council created ONUC to help maintain order, offer technical assistance, and oversee the withdrawal of Belgian troops. ONUC's mandate was later broadened: it was to prevent civil war from breaking out, reunite the country, and prevent foreign military and civilian personnel from getting involved in the crisis outside the United Nations' purview. The mission was concluded in June 1964.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 21 mm central band of dark green flanked by stripes of 2 mm of white and 4 mm of dark blue.
Other: Originally, this medal was hung from the same ribbon used for the UNTSO and UNOGIL medals, with the addition of a small bronze CONGO bar. This was replaced in 1963 by the current ribbon.
Number: 1,900.
United Nations Temporary Executive Authority (UNTEA) in West New Guinea, 1962–63
UNTEA Medal.
Origins: In August 1962 an agreement was reached whereby the Netherlands would transfer control over West New Guinea to Indonesia. During the transition period, the territory was to be administered by the United Nations, which established UNTEA in October of that year. UNTEA and a U.N. Security Force were given the task of ensuring an orderly transfer. The mission was completed in April 1963.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide U.N. blue ribbon with three 3 mm stripes in the centre of dark green, white, and light green.
Number: 13.
United Nations Yemen Observer Mission (UNYOM), 1963–64
UNYOM Medal.
Origins: After a coup in Yemen in July 1963, this mission was established to oversee and certify that the disengagement agreement between Saudi Arabia and the United Arab Republic was put into practice. The mission ended in September 1964.
Criteria: Sixty consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon shaded brown and bordered on each side by 2 mm of U.N. blue.
Number: 30.
United Nations Peacekeeping Force in Cyprus (UNFICYP), 1964–Present
UNFICYP Medal.
Origins: This peacekeeping force was organized in March 1964 to prevent violence from breaking out between Greek and Turkish Cypriots. After a military coup on the island, Turkey invaded northern Cyprus. UNFICYP's mandate was expanded to include supervision of the U.N.-sponsored ceasefire and occupation of a buffer zone between the Greek Cypriot territory to the south and that of the Turkish Cypriots in the north.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a central white 9 mm stripe bordered on each side by 1 mm of dark blue.
Number: 35,000.
United Nations India-Pakistan Observer Mission (UNIPOM), 1965–66
UNIPOM Medal.
Origins: UNIPOM was established in September 1965 after fighting broke out between India and Pakistan during the summer of that year. The mission's observers were dispatched to the border (except within the state of Jammu and Kashmir, which was UNMOGIP's responsibility) to oversee a ceasefire as well as the return of troops from both sides to the lines before August 5, 1965. The mission ended in March 1966.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: The same as that issued for UNMOGIP.
Number: 112.
United Nations Emergency Force Middle East (UNEFME), 1973–79
UNEFME Medal.
Origins: Also known as United Nations Emergency Force II, UNEFME was set up in October 1973 to uphold the ceasefire between Egypt and Israel after the Yom Kippur War and to occupy the buffer zones established under agreements made between the two sides on January 18, 1974, and September 4, 1975. The mission concluded in July 1979.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with 9 mm stripes of U.N. blue at either edge and down the centre a 2 mm stripe of sand brown flanked by 1 mm stripes of blue and additional 6 mm bands of sand brown.
Number: 11,500.
United Nations Disengagement Observer Force (UNDOF), Golan Heights, 1974–Present
UNDOF Medal.
Origins: Established in 1974 to supervise the ceasefire between Israel and Syria and the disengagement of the two countries' forces as well as to establish a buffer zone to be patrolled by the United Nations.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 1 mm maroon stripe in the centre flanked on each side by 4 mm of sky blue, 1 mm of black, 3 mm of white, and 9 mm of maroon.
Number: 12,425.
United Nations Interim Force in Lebanon (UNIFIL), 1978–Present
UNFIL Medal.
Origins: Established in March 1978 after Israel's invasion of Lebanon, this mission's task was to confirm the withdrawal of Israeli forces from southern Lebanon, to restore stability between the two nations, and to help restore the Lebanese government's authority over its territory.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with 9 mm edges of U.N. blue and a central 11 mm stripe of green flanked on each side by 1 mm stripes of white, red, and white.
Number: 117.
United Nations Iran-Iraq Military Observer Group (UNIIMOG), 1988–91
UNIIMOG Medal.
Origins: Established in August 1988 after the Iran-Iraq War to oversee the ceasefire and withdrawal of all forces to within the respective nations' internationally recognized boundaries. The mission ended in February 1991.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 17 mm central sripe of U.N. blue edged on each side by 3 mm of red and 3 mm of white, with 3 mm of green on the left and 3 mm of black on the right.
Number: 581.
United Nations Transitional Assistance Group (UNTAG), Namibia, 1989–90
UNTAG Medal.
Origins: Established in April 1989 to ensure that Namibia's transition to independence from South Africa was orderly and that free and fair elections under the supervision and control of the United Nations were held. This mission concluded in March 1990.
Criteria: Ninety days of consecutive service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon edged on both sides with 8 mm stripes of U.N. blue with a 17 mm central stripe of sand brown containing 1 mm stripes of black, yellow, red, green, and royal blue.
Number: 401.
United Nations Observer Group in Central America (ONUCA), 1989–92
ONUCA Medal.
Origins: Established in December 1989 to ensure that the governments of Costa Rica, El Salvador, Guatemala, Honduras, and Nicaragua had ceased to provide aid to irregular (i.e., paramilitary) forces and insurrectionists and that none of these countries was being used to launch attacks on any of the others. The mission ended in January 1992.
Criteria: Ninety days of consecutive service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon edged with 6 mm of royal blue and 7 mm of U.N. blue, the centre divided by alternating 1 mm stripes of green and white.
Number: 350.
United Nations Iraq-Kuwait Observer Mission (UNIKOM), 1991–2003
UNIKOM Medal.
Origins: UNIKOM was established in April 1991 to monitor a demilitarized zone along the Iraq-Kuwait boundary and to patrol the Khor Abdullah waterway.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide sand brown ribbon with a central 4 mm stripe of U.N. blue.
Number: 605.
United Nations Angola Verification Mission (UNAVEM), 1988–97
UNAVEM Medal.
Origins: Originally established in December 1988 to ensure that Cuban troops were withdrawn from Angola, UNAVEM's mandate was altered in May 1991. Now known as UNAVEM II, its new role was to monitor the ceasefire called for under the peace accords that had been signed and to supervise elections. With the renewal of fighting between government troops and União Nacional para a Independência Total de Angola (UNITA) rebels, UNAVEM entered a third phase (UNAVEM III) during which it was tasked with demobilizing the UNITA forces, restoring order, and overseeing new elections. The mission was concluded in June 1997.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 9 mm central stripe of U.N. blue bordered on either side by 1 mm of black, 2 mm of white, 2 mm of red, and 8 mm of of yellow.
Number: 60.
United Nations Mission for the Referendum in Western Sahara (MINURSO), 1991–Present
MINURSO Medal.
Origins: Established in May 1991 to oversee a referendum through which the people of Western Sahara were to choose between integration with Morocco or independence.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 27 mm central stripe of sand brown edged on each side by 4 mm of U.N. blue.
Number: 140.
United Nations Observer Mission in El Salvador (ONUSAL), 1991–95
ONUSAL Medal.
Origins: Established in May 1991 to monitor agreements ending the civil war between the government of El Salvador and the Frente Farabundo Martí para la Liberación Nacional (FMLN). This mission was expanded in January 1992 to include monitoring the ceasefire and supervising the demobilization of the warring forces as well as maintaining order during the period while the national police force was phased out and replaced by a new civilian police force. ONUSAL also supervised the 1994 elections. Upon completing its tasks, the mission was concluded in April 1995.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 7 mm stripe of white bordered on each side by 7 mm of dark blue and 7 mm of U.N. blue.
Number: 55.
United Nations Protection Force in the Former
UNPROFOR Medal.
Yugoslavia (UNPROFOR), 1992–95
Origins: Established on an interim basis in February 1992 to create a stable environment in which an overall settlement of the Yugoslavian crisis could be negotiated. U.N. troops were assigned to ensure that areas designated as "U.N. Protected Areas" became and remained demilitarized and to remove the threat of armed attack in these regions. U.N. police monitors were assigned to supervise local police forces, primarily with an eye toward preventing human rights abuses. UNPROFOR also helped the humanitarian agencies of the United Nations with the return of all displaced persons who so desired. The mission's mandate was extended several times: to reopen the Sarajevo airport for humanitarian purposes, to establish a security zone around Sarajevo and its airport, to protect convoys of released detainees in Bosnia and Herzegovina, to monitor arrangements for the complete withdrawal of the Yugoslavian Army from Croatia, to oversee the demilitarization of the Prevlaka peninsula and the removal of heavy weapons from neighbouring areas of Croatia and Montenegro, to monitor compliance with a ban on military flights, and to establish a U.N. presence in Macedonia. UNPROFOR also monitored a ceasefire between the Bosnian government and Bosnian-Croat forces in February 1994, and another between the Bosnian government and Bosnian Serb forces, which became effective on January 1, 1995. On March 31, 1995, UNPROFOR was replaced with three separate but interrelated peacekeeping operations: United Nations Confidence Restoration Operation in Croatia (UNCRO) and United Nations Preventive Deployment Force (UNPREDEP), both under the auspices of United Nations Peace Forces (UNPF). In January 1996 the UNPF mission was phased out.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 9 mm stripe of red bordered on each side by 1 mm of white, 3 mm of U.N. blue, and 6 mm of green on the left and 6 mm of brown on the right, each side edged in 3 mm of U.N. blue.
Number: 6,650.
United Nations Advanced Mission in Cambodia (UNAMIC), 1991–92
UNAMIC Medal.
Origins: Established in October 1991 to liaise between the four warring factions in Cambodia in matters related to the ceasefire and to launch a mine-awareness training program. UNAMIC's mandate was later broadened to include training in mine clearance and to launch a program to remove land mines. The mission's mandate expired in March 1992 when the United Nations Transitional Authority in Cambodia was established.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 3 mm stripe of white bordered on each side by 4 mm of dark blue, 1 mm of yellow, 4 mm of red, and 7 mm of U.N. blue.
Number: 7.
United Nations Transitional Authority in Cambodia (UNTAC), 1992–93
UNTAC Medal.
Origins: This mission succeeded UNAMIC. Its role was to supervise the ceasefire agreed to in October 1991, to oversee the withdrawal of foreign forces and the demobilization of armed forces from the four parties, to temporarily assume civil administrative duties, to ensure that the police and other bodies respected human rights, to repatriate and resettle refugees and displaced persons, and to organize and supervise elections. After the elections, the mission was terminated on November 15, 1993, its place to be taken by the United Nations Military Liaison Team (UNMLT). Consisting of twenty military observers, this team's six-month mandate was to monitor the situation, help the government deal with military loose ends related to the Paris peace agreement, and to report to the secretary general on matters affecting security in Cambodia. UNMLT was established for a single period of six months. UNMLT military observers are eligible to receive the UNTAC Medal.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 3 mm central stripe of white bordered on each side by 1 mm of red, 4 mm of U.N. blue, 1 mm of dark blue, and 10 mm of dark green.
Number: 465.
United Nations Operation in Somalia (UNOSOM), 1992–93
UNOSOM Medal.
Origins: UNOSOM was established in April 1992. Fifty unarmed uniformed military observers were dispatched to Mogadishu to monitor the ceasefire there. They were to be stationed along the line that divided the Somali capital into two zones. The mission was also to protect distribution centres and the port of Mogadishu and to escort supply convoys throughout the vicinity. They were also to provide security for U.N. personnel, equipment, and supplies at the airport in Mogadishu. On August 28 the Security Council authorized the posting of four more U.N. security units to UNOSOM to protect humanitarian aid convoys and distribution centres throughout Somalia.
As ongoing fighting interfered with relief efforts, the Security Council on December 3 authorized UNOSOM and UNITAF — the Unified Task Force spearheaded by the United States — to work together using any means necessary to ensure a stable environment for humanitarian and relief efforts to operate. Once UNITAF's task was accomplished, military command was assumed by the United Nations. Meanwhile, UNOSOM remained in charge of its original political and humanitarian mandates. In March 1993, UNOSOM II was created. Like UNITAF, it was authorized to use force if necessary to create a secure environment to distribute humanitarian aid. In February 1994, after several violent incidents and attacks on U.N. soldiers, UNOSOM II's mandate was updated to exclude the use of coercion. Unable to make satisfactory progress, UNOSOM II was withdrawn in early March 1995.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with an 11 mm central stripe of U.N. blue bordered on each side by 2 mm of olive green and 10 mm of sand brown.
Number: 343.
United Nations Operation in Mozambique (ONUMOZ), 1992–94
ONUMOZ Medal.
Origins: Established in October 1992, ONUMOZ's mandate was to monitor the ceasefire, the withdrawal of foreign forces, and the disbandment of unofficial armed forces; to secure transportation routes; to monitor elections; and to assist humanitarian aid operations. Having fulfilled its mandate, the mission was wound down in January 1995.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central stripe of 15 mm U.N. blue bordered on each side by 5 mm of white and 5 mm of green.
Number: 45.
United Nations Observer Mission in Uganda-Rwanda (UNOMUR), 1993–94
UNOMUR Medal.
Origins: Established in June 1993, UNOMUR's mission was to monitor the border between Uganda and Rwanda and ensure that no military matériel — primarily in the form of weapons and ammunition — reached Rwanda. The mission was concluded in October 1994.
Criteria: Service of 180 consecutive days.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 9 mm stripe of U.N. blue bordered on each side by 1 mm of white, 4 mm of black, 4 mm of yellow, and 4 mm of red.
Number: 5.
United Nations Assistance Mission in Rwanda (UNAMIR), 1993–96
UNAMIR Medal.
Origins: Established in October 1993, UNAMIR's mandate was tasked with monitoring the ceasefire agreement, helping to secure the capital of Kigali, helping to remove land mines, coordinating humanitarian aid efforts, and protecting and repatriating refugees. UNAMIR also helped to protect human rights workers and the personnel of the International Tribunal for Rwanda and assisted in the establishment of a new national police force. UNAMIR's mandate concluded in March 1996.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. medal.
Ribbon: A 35 mm wide ribbon with a central 9 mm stripe of U.N. blue edged on each side by 1 mm of white, 4 mm of black, 4 mm of green, and 4 mm of red.
Number: 375.
United Nations Missions in Haiti (UNMIH, UNSMIH, UNTMIH, MIPONUH, MICAH), 1993–Present
UNMIH Medal.
Origins: This U.N. medal issue is perhaps the most complex, there being five separate bars and missions.
* • UNMIH: Originally established on September 23, 1993, with the original task of helping to modernize Haiti's armed forces and to establish a new police force. Haitian military authorities did not co-operate and prevented the mandate from being carried out. In 1995 a renewed UNMIH took over from the U.S.-led multi-national force to maintain order, help create a civilian police force, and oversee elections. Its mission ended in June 1996.
* • UNSMIH: UNMIH was succeeded in July 1996 by the United Nations Support Mission in Haiti (UNSMIH). Its mandate expired on July 31, 1997.
* • UNTMIH. The United Nations Transition Mission in Haiti was the third U.N. peacekeeping operation in Haiti. It existed from July 30, 1997, until November 30, 1997. Its role was to train the Haitian National Police (HNP).
* • MIPONUH: In December 1997, UNTMIH was succeeded by the United Nations Civilian Police Mission in Haiti (MIPONUH). Its mandate was to continue training the HNP.
* • MICAH: MIPONUH was succeeded by the International Civilian Support Mission in Haiti (MICAH) on March 16, 2000. The establishment of MICAH was approved by the U.N. General Assembly in Resolution A/54/193 of December 17, 1999. Its mandate is to reinforce the results its predecessors had yielded. It was also assigned to promote human rights and support the police and justice system.
UNMIH Medal bars.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 7 mm blue and 7 mm red stripe in the centre bordered on each side by 1 mm of white and 7 mm of U.N. blue.
Number: Single Bars: UNMIH, 902; UNSMIH, 820; MIPONUH, 102; UNTMIH, 112. Multiple Bars: UNMIH and UNSMIH, 756; UNSMIH and UNTMIH, 815[2]
United Nations Verification Mission in Guatemala (MINUGUA), 1994–97
MINUGUA Medal.
Origins: Created in September 1994 to verify the establishment of human rights in Guatemala. The mission was concluded in May 1997.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central stripe of U.N. blue flanked on each side by stripes of dark blue, white, green, white, and dark blue.
Number: 64.
United Nations Mission in the Central African Republic (MINURCA), 1998–2000
MINURCA Medal.
Origins: After social unrest and army mutinies took place in the Central African Republic, MINURCA was formed to monitor the area and ensure that the truce agreements were put into operation. The mission was concluded in February 2000.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 5 mm central stripe of U.N. blue bordered on each side by five 3 mm stripes of yellow, green, red, white, and dark blue.
Number: 143.
United Nations Preventative Deployment Force in Macedonia (UNPREDEP), 1995–99
UNPREDEP Medal.
Origins: UNPREDEP was created in March 1995 to carry out the work of UNPROFOR within the Former Yugoslav Republic of Macedonia. Its mandate was to monitor the borders and report any potential external threats to the country's stability. The mission was concluded in February 1999.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central stripe of 5 mm of red bordered on each side by 1.5 mm of yellow, 2 mm of red, 1.5 mm of yellow, 2 mm of red, 1 mm of white, and 7 mm of U.N. blue.
Number: 20 .
United Nations Mission in Bosnia and Herzegovina (UNMIBH), 1995–2002
UNMIBH Medal.
Origins: UNMIBH was established in December 1995 to carry out a variety of law-enforcement functions and to create favourable conditions for elections to be held. The mission was concluded in December 2002.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 6 mm stripe of white bordered on either side by 7 mm of U.N. blue and 7 mm of green (left) or 7 mm of red (right).
Number: 250.
United Nations Mission of Observers in Prevlaka (UNMOP), Croatia, 1996–2002
UNMOP Medal.
Origins: Established in January 1996, UNMOP was mandated with a mission to monitor the demilitarization of the Prevlaka Peninsula in Croatia. The mission was concluded in December 2002.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a central 5 mm yellow stripe bordered on each side by 1 mm of white, 4.5mm of dark blue, 4.5mm of U.N. blue, and 5 mm of dark blue.
Number: 27.
United Nations Interim Administration Mission in Kosovo (UNMIK), 1999–Present
UNMIK Medal.
Origins: UNMIK was established on June 10, 1999, to act as a civil authority in Kosovo, to ensure the safe return of refugees, and to prevent conflicts between parties from escalating.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 15 mm central stripe of dark blue bordered on each side by 1 mm of white and 9 mm of U.N. blue.
Number: 155.
United Nations Mission in Sierra Leone (UNOMSIL/UNAMSIL), 1998–2005
UNOMSIL/UNAMSIL Medal.
Origins: UNOMSIL was established in June 1998 to disarm and demobilize the former combatants in Sierra Leone's lengthy civil war. In October 1999 the United Nations established UNAMSIL, a much larger mission (with a maximum of six thousand military personnel, including 260 military observers) designed to help enact the Lomé Peace Agreement. At the same time UNOMSIL was cancelled.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with an 11 mm central stripe of U.N. blue bordered on either side by 4 mm each of green, white, and dark blue.
Number: 8 .
United Nations Mission in East Timor and United Nations Transitional Administration in East Timor (UNAMET and UNTAET), 1999–2002
UNAMET/UNTAET Medal.
Origins: UNAMET was established in June 1999 to organize and conduct a referendum on whether the people of East Timor wished to separate from Indonesia. When the people voted for independence, violence broke out. In September, UNTAET was organized and dispatched to provide security and maintain order, develop civil institutions, and lay the groundwork for self-government.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 7 mm central stripe of white bordered on either side by 2 mm of yellow, 2 mm of red, and 10 mm of U.N. blue.
Number: Approximately 30.
United Nations Organization Mission in the Democratic Republic of the Congo (MONUC), 1999–2010
MONUC Medal.
Origins: The primary assignment of this mission, established on November 30, 1999, is to liaise between the armed forces of all six parties to the Lusaka Ceasefire Agreement signed in July 1999, to investigate violations of the agreement, and to aid in the release of all prisoners of war.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 9 mm central stripe of dark blue bordered on either side by 2 mm of yellow and 11 mm of U.N. blue.
Number: 22.
United Nations Mission in Ethiopia and Eritrea (UNMEE), 2000–08
UNMEE Medal.
Origins: This mission was created on September 15, 2000. Its primary task is to monitor the ceasefire between Ethiopia and Eritrea.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 2 mm central stripe of green bordered on either side by 6.5mm of sand brown and 10 mm of U.N. blue.
Number: 6.
United Nations Stabilization Mission in Haiti (MINUSTAH), 2004–Present
MINUSTAH Medal.
Origins: The United Nations has been active in Haiti since 1994 in a variety of missions. In February 2004, forces opposed to President Jean-Bertrand Aristide marched on Port-au-Prince and overthrew the president. An interim government took power and requested that the United Nations establish an interim force to help restore peace. Under resolution 1542 of April 30, 2004, the U.N. Security Council established MINUSTAH. The mission's mandate is to create a secure environment, to promote human rights, and to assist in the delivery of humanitarian aid.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a 2 mm central stripe of U.N. blue bordered on each side by 5.5mm stripes of white, green, and dark blue.
Number: 1,700.
United Nations Operation in Ivory Coast (ONUIC), 2004–Present
ONUIC Medal.
Origins: The Ivory Coast has been enmeshed in crisis for the past decade, suffering two coups d'état — one in 1999 and another in 2001, followed by the outbreak of civil war in 2002. A U.N.-sponsored ceasefire was established in early 2004. Under U.N. Security Council Resolution 1528 of February 27, 2004, ONUIC was established to monitor the ceasefire and movements of armed groups; support disarmament, demobilization, reintegration, repatriation, and resettlement; protect U.N. personnel, institutions, and civilians; support humanitarian assistance and the implementation of the peace process; provide assistance in the field of human rights; restore civilian policing in the country; and re-establish the authority of the judiciary and the rule of law throughout the country.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a 3 mm central stripe of white bordered on each side by 3 mm of orange (left) and green (right).
Number: 122.
United Nations Mission in Sudan (UNMIS), 2005–11
UNMIS Medal.
Origins: The Sudan has been wracked by civil war almost from the time of its independence. The most recent civil war commenced in 1983 with a ceasefire being achieved in 2005. To facilitate the attainment of peace and the restoration of civil society the United Nations passed Security Council Resolution 1590 of March 24, 2005, that established UNMIS, which was established to support the implementation of the Comprehensive Peace Agreement and to perform functions relating to humanitarian assistance and the protection and promotion of human rights.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a central 3 mm stripe of dark blue bordered on each side by 3 mm of white and 3 mm of dark blue.
Number: 625.
United Nations Integrated Mission in Timor-Leste (UNMIT), 2006–12
UNMIT Medal.
Origins: This mission was established in the newly founded Democratic Republic of Timor-Leste, which was previously known as East Timor. The former Portuguese colony had been seized by Indonesian forces in 1975 and made into that country's twenty-seventh province. In 1999 the United Nations oversaw Indonesia's withdrawal from the territory with the United Nations Transitional Administration in East Timor, and the democratic republic was established as an independent state on May 20, 2002. In August 2006 the United Nations established UNMIT to quell factional fighting among key ethnic groups, to assist in the consolidation of stability, to foster civil society, and to create greater social cohesion. This mission ended on December 31, 2012.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a 9 mm central stripe of red bordered on each side by 2 mm strips of yellow, black, and white.
Number: 63.
United Nations African Union Hybrid Mission in Darfur (UNAMID), 2007–Present
UNAMID Medal.
Origins: This mission, undertaken in tandem with the African Union, was created by U.N. Security Council Resolution 1769 of 2007. It was established in Darfur, Sudan, and replaced the African Union Mission in Sudan (AMIS). The principal goal of the U.N. mission is to support the implementation of the Darfur Peace Agreement and to prevent armed attacks and protect civilians in Darfur. The United Nations has also provided security for humanitarian aid and assistance and promotes human rights and the rule of law.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a 7 mm central stripe of white flanked on each side by 3 mm stripes of yellow and green.
Number: 96.
United Nations Mission in the Republic of South Sudan (UNMISS), 2011–Present
UNMISS Medal.
Origins: After the creation of the Republic of South Sudan on July 8, 2011, UNMISS was established to provide support for peace consolidation, to foster the development of a civil society, to encourage economic development, and to offer support for the newly created country. This mission has a focus on conflict prevention and resolution with an overall purpose of strengthening South Sudan by aiding in the establishment of the rule of law.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue with a central dark green stripe bordered on each side by white and U.N. blue stripes, each stripe being separated by thin black stripes.
Number: 147.
United Nations Special Service Medal (UNSSM), 1974–Present
UNSSM Medal.
Origins: Created to recognize military and civilian police personnel who have served the United Nations in capacities other than in established peacekeeping missions or at U.N. Headquarters.
Criteria: Ninety consecutive days of service.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon with a broad central stripe of U.N. blue bordered on either side by 5 mm stripes of white.
Other: Awards can be made for service dating back to 1974. The United Nations approved this medal in June 1995. To identify the theatre in which the medal was earned, a bar is sometimes added with the name of the country or the U.N. organization (UNHCR, UNSOM, et cetera). These bars are not created centrally by the United Nations but locally by mission commanders and therefore their design and quality vary.
Number: 247.
United Nations Headquarters Service (UNHQ), 1974–Present
UNHQ Medal.
Origins: Created on June 16, 1997, to recognize military personnel who hold staff positions at U.N. Headquarters in New York City dating back to 1974.
Criteria: Ninety consecutive days of service at U.N. Headquarters.
Insignia: Standard U.N. Medal.
Ribbon: A 35 mm wide ribbon of U.N. blue.
Number: 235.
# 25
The North Atlantic Treaty Organization Medals
As a founding member of the North Atlantic Treaty Organization (NATO), Canada has played a small but significant role in the organization since its iestablishment in 1948. Beginning in the early 1990s when civil war broke out in the Federal Socialist Republic of Yugoslavia, NATO has increasingly become involved in peacemaking and peacekeeping missions. Recently, NATO has assisted in Afghanistan, Sudan, and Iraq in a variety of training and logistical support roles.
The vast majority of these missions have been sponsored and carried out under a resolution of the U.N. Security Council, part of a growing trend for peacekeeping and peacemaking operations to be sanctioned by the United Nations, yet carried out by organizations such as NATO.
The ten NATO medals that have been awarded to Canadians and accepted by the Canadian government are similar to the various U.N. peacekeeping medals in that the design of all ten is uniform and they are differentiated only by their ribbons and in some cases by the bars awarded with them. A NATO medal for the International Security Assistance Force (ISAF) mission in Afghanistan also exists but was declined by Canada and cannot be received by Canadians because this service is already recognized by Canadian medals such as the General Campaign Star or General Service Medal — South-West Asia.
NATO made two significant changes to its medal policy. In 2003 it changed its first three medals — which had distinct ribbons and bars — to a system of Article 5 and Non-Article 5 medals based on whether the mission was triggered by this article of the NATO Treaty (an attack against a NATO country is deemed an attack against all). The former would have ARTICLE 5 bars and ribbons bearing white and gold stripes increasing in number with each mission, while the latter would have NON-ARTICLE 5 bars and ribbons bearing white and silver stripes increasing in number with each mission. That same change also abolished the use of numerals for multiple tours and cancelled eligibility for the "adjacent areas."
In 2011 a further review was conducted and all medals still on issue were standardized so that all Article 5 medals would bear a specific operation bar but share the same ribbon with two white and gold stripes and Non-Article 5 medals would also have specific bars and share a ribbon with two white and silver stripes. Inasmuch as the ribbons would be similar in each category, devices in the form of a miniature bar would now be worn on the undress ribbon for differentiation. Furthermore, the criteria were amended from thirty days of service, either consecutive or cumulative, to thirty consecutive or sixty cumulative days. New tour numerals, this time appearing on a little bronze square, were also reintroduced to recognize multiple tours.
NATO medals have been manufactured by Eekellers-Centini of Belgium. They are issued in a rectangular blue plastic case; some issues bear the NATO star, while others are plain. There are no postnominals associated with any of these honours.
NATO Service Medals
Insignia: The obverse of the standard NATO Service Medal is bronze, 36 mm in diameter, and bears the NATO star surrounded by a wreath of olive leaves. The reverse bears the motto IN THE SERVICE OF PEACE AND FREEDOM, a laurelled bar, and AU SERVICE DE LA PAIX ET DE LA LIBERTÉ encircled by NORTH ATLANTIC TREATY ORGANIZATION — ORGANISATION DU TRAITÉ DE L'ATLANTIQUE NORD.
Suspender: A bronze ring passes through the top of the medal through which the ribbon passes.
Bars: Each NATO medal is issued with an operation/theatre-specific bar.
Numerals: A small bronze numeral is worn on the ribbon for additional qualifying periods of service, and a number of bars have also been issued to specific medals. In January 2011 a square bronze numeral was created to denote multiple tours on the following NATO missions: Article 5 Active Endeavour, Non-Article 5 for Operations in the Balkans, Non-Article 5 NATO Training Mission in Iraq, Non-Article 5 Logistical Support to the African Union in the Sudan, Non-Article 5 Operations and Activities Related to Africa, and Non-Article 5 Operation Unified Protector, Libya.
Naming: None.
The NATO Medal, Former Yugoslavia
NATO Medal, Former Yugoslavia obverse.
Origins: After Slovenia, Croatia, and Macedonia seceded from the Federal Socialist Republic of Yugoslavia, Bosnia and Herzegovina also voted late in 1991 for independence from Yugoslavia, but much of the Bosnian Serb population, which represented a minority of the new republic's population, refused to separate. A bloody civil war commenced. In 1995, NATO became involved in the Bosnian war, initially in support of the United Nations along with the European Union. The alliance monitored and enforced U.N. sanctions, enforced the no-fly zone over Bosnia, and provided support for the U.N. Protection Force (UNPROFOR). NATO also played a key role in carrying out air strikes that ended the siege of Sarajevo.
In December 1995 the governments of Bosnia, Croatia, and Serbia were brought together in Dayton, Ohio, to negotiate, and a peace treaty was signed. On December 20 a NATO-led multinational force — the Implementation Force (IFOR) — was deployed in Bosnia and Herzegovina to bring about and maintain an end to hostilities and to separate the armed forces of Bosnia's two newly created entities: the Federation of Bosnia and Herzegovina and the Republika Srpska. These goals were attained by June 1996. Later that year IFOR was succeeded by the Stabilization Force (SFOR), which was also deployed as part of the Dayton Peace Accord. This peacekeeping operation, twenty thousand strong, is assigned to maintain and enforce security to "enable the country to rebuild after the devastation of years of conflict." SFOR is composed of sixteen NATO members and thirteen partner countries.
NATO Medal, Former Yugoslavia reverse (standard to all NATO medals).
Because this was the first large-scale NATO peacekeeping operation, it was decided to issue a service medal similar to those the United Nations awarded to peacekeepers. The NATO medal for the former Yugoslavia was the first NATO medal instituted and was created by the alliance's secretary general on December 20, 1994.
Criteria: Thirty consecutive or cumulative days of service on the land, at sea, or in the airspaces of the former Yugoslavia, Albania, the former Republic of Macedonia, or the Adriatic Sea (the theatre of operations), or ninety days of continuous or accumulated service in the territories of Italy, Greece, Hungary, and Austria in direct support of the NATO operations in the former Yugoslavia (the adjacent area). The service must have been rendered between July 1, 1992, and December 31, 2002.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 4 mm white stripe set 4 mm from each edge.
Bar: The medal bears a plain bronze bar with FORMER YUGOSLAVIA or EX-YOUGOSLAVIE in bas-relief. A bronze numeral 2 is worn on the ribbon of the medal for an additional 180 days of service in the theatre of operation or 540 days of service in the adjacent area as defined above. Further service would be recognized by a numeral 3 and so forth. Tour numerals are worn below the bar.
Other: This medal was replaced by the NATO Non–Article 5 Medal for operations in the Balkans on January 1, 2003.
Number: More than 12,000.
The NATO Medal, Kosovo
NATO Medal, Kosovo obverse.
Origins: Throughout the 1990s the situation in the Balkan states — in what was formerly the Federal Socialist Republic of Yugoslavia — deteriorated significantly. In 1998 protests erupted in Kosovo and Serbian police reacted with excessive force, while members of the Kosovo Liberation Army (KLA) began a series of terrorist raids on Serb targets and civilians. More than two hundred and fifty thousand residents of Kosovo were displaced by the violence, and the United Nations passed a series of resolutions designed to yield a political solution. The government of the Federal Republic of Yugoslavia (essentially Serbia) refused to cease its actions, and NATO, authorized by the United Nations, undertook Operation Allied Force — an extensive bombing campaign that struck targets throughout Yugoslavia. In June 1999, after eleven weeks of bombing, the Yugoslavian government relented and agreed to allow an International Security Force (KFOR) into Kosovo.
Shortly thereafter, a U.N. resolution called for the deployment of an effective international civil and security presence in Kosovo, which represented an official endorsement of the KFOR mission. The mission itself had many objectives, though it was primarily charged with establishing a lasting end to hostilities.
Criteria: Thirty consecutive or cumulative days of service on the land, at sea, or in the airspace of Kosovo and other territories of the Federal Republic of Yugoslavia, Albania, the former Republic of Macedonia, and the Adriatic and Ionian Seas (the theatre of operations), or ninety days of continuous or accumulated service in the territories of Italy, Greece, and Hungary in direct support of the NATO operations in Kosovo (the adjacent area). Aircrew who flew fifteen sorties in the airspace of Kosovo and the other territories of the Federal Republic of Yugoslavia during Operation Allied Force between March 24 and June 10, 1999, also qualify. Aircrew who completed fewer than fifteen sorties may combine the number of completed sorties with other qualifying service between October 13, 1998, and December 31, 2002, to qualify.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 6 mm white stripe in the centre and edged on either side with 2 mm of white.
Bar: A plain bronze bar with KOSOVO in bas-relief. A bronze numeral 2 is worn on the ribbon of the medal for an additional 180 days of service in the theatre of operation or 540 days of service in the adjacent area as defined above. Further service would be recognized by a numeral 3. Tour numerals are worn below the bar.
Other: This medal was replaced by the NATO Non-Article 5 Medal for the Balkans on January 1, 2003. Those who qualified for this medal for service with Operation Allied Force may now exchange it for the General Campaign Star or General Service Medal — Allied Force.
Number: 2,200.
The NATO Medal, Former Yugoslav Republic of Macedonia
NATO Medal, Former Yugoslav Republic of Macedonia obverse.
Origins: After the disintegration of the Federal Socialist Republic of Yugoslavia in the early 1990s, Macedonia declared its independence. In 1993 it joined the United Nations under the name "the Former Yugoslav Republic of Macedonia." In the wake of the influx of Albanian refugees from Kosovo in 1999, Macedonia's ethnic Albanian population began to lobby for more rights and a constitutional position. Initially, the Macedonian government responded coldly to these requests, and in early 2001 the National Liberation Army (NLA), an ethnic Albanian group closely related to the Kosovo Liberation Army, launched a mortar attack on a Macedonian police station. The conflict escalated, and Macedonian security forces engaged the NLA throughout early 2001. A tentative political solution was arrived at in August 2001, and President Trajkovski of Macedonia asked NATO troops to disarm the NLA and other ethnic Albanian groups. Operation Essential Harvest was launched on August 22, 2001, and began in earnest on August 27. Approximately 3,500 NATO troops engaged to disarm the NLA. The operation lasted until September 26, 2001.
Criteria: Twenty-five consecutive or cumulative days of service on the land or in the airspace of the Former Yugoslav Republic of Macedonia between June 1 and August 27, 2001, or between August 27 and September 27, 2001, or thirty consecutive or cumulative days of service between August 27, 2001, and December 31, 2002.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, edged with 3 mm of white, a pair of 4 mm white stripes superimposed on the blue and set 7 mm from either edge.
Bars: None.
Number: 297.
The Article 5 NATO Medal for Operation Eagle Assist
Article 5 NATO Medal for Operation Eagle Assist obverse.
Origins: After the terrorist attacks on the United States on September 11, 2001, NATO launched Operation Eagle Assist under which the alliance was to assist the United States and demonstrate its resolve against terrorism. Deployment was authorized on October 12, 2001, and was completed on May 16, 2002.
Criteria: Thirty consecutive or sixty cumulative days of service in the area of operation between October 12, 2001, and May 16, 2002. The area of operation was defined as North American airspace for which the North American Aerospace Defense Command (NORAD) was responsible.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a central 5 mm white stripe superimposed with a thin gold thread.
Bars: The medal bears a plain bronze bar with ARTICLE 5 in bas-relief.
Number: 60.
The Article 5 NATO Medal for Operation Active Endeavour
Article 5 NATO Medal for Operation Active Endeavour obverse.
Origins: After the terrorist attacks on the United States on September 11, 2001, NATO launched Operation Active Endeavour, the purpose of which was to provide a NATO presence and demonstrate the alliance's resolve in the eastern Mediterranean Sea. A standing naval force was deployed on October 26, 2001. This mission is ongoing.
Criteria: Thirty consecutive or sixty cumulative days of service in the international waters of the Mediterranean Sea and the airspace above it.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 5 mm white stripe inset 5 mm from each edge and bisected by a single thin gold thread.
Bars: Originally awarded with a plain bronze bar with ARTICLE 5 in bas-relief and with the undress ribbon unadorned. Medals issued since January 1, 2011, bear a bar with the inscription ACTIVE ENDEAVOUR and wear a miniature device of the same design on the undress ribbon.
Number: 260.
The Non-Article 5 NATO Medal for Operations in the Balkans
Origins: This medal replaces the NATO Medal for the Former Yugoslavia, the NATO Medal for Kosovo, and the NATO Medal for the Former Yugoslav Republic of Macedonia. It was introduced to reduce the number of NATO medals awarded for service in the Balkans. This medal was first issued on January 1, 2003, at the same time as the other three were phased out.
Criteria: Thirty consecutive or sixty cumulative days of service in the area of operation (the Balkans Joint Operational Area), commencing January 1, 2003. Service in adjacent areas no longer qualifies as it had under previous NATO medals for service in the Balkans.
Insignia: Standard NATO Service Medal.
Ribbon: Initially, a NATO blue, 36 mm wide, with a central 4 mm white stripe superimposed with a 1 mm silver thread. Medals issued from January 1, 2011, have 2 5 mm white stripes inset 5 mm from each edge, each bisected by a single thin silver thread.
Non-Article 5 NATO Medal for Operations in the Balkans.
Bars: Originally awarded with a plain bronze bar with NON ARTICLE 5 in bas-relief and with the undress ribbon unadorned. Medals issued since January 1, 2011, bear a bar with the inscription BALKANS and wear a miniature device of the same design on the undress ribbon.
Number: Approximately 500.
The Non-Article 5 NATO Medal for Service with the NATO Training Implementation Mission and the Nation Training Mission in Iraq (NTM-IRAQ)
Non-Article 5 NATO Medal for Service with the NATO Training Implementation Mission and National Training Mission in Iraq (NTM-IRAQ) obverse.
Origins: After the U.S.-led campaign to remove Saddam Hussein as the leader of Iraq, NATO undertook the task to train the new Iraqi Armed Forces through its NATO Training Implementation Mission (NTIM-I) and NATO Training Mission (NTM-I) in Iraq.
Criteria: Thirty consecutive or sixty cumulative days of service in the area of operation (Iraq) from August 18, 2004, to December 31, 2011.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 5 mm white stripe inset 5 mm from each edge bisected by a single thin silver thread.
Bars: The medal bears a plain bronze bar with NTM — IRAQ centred and in bas-relief.
Number: Approximately 500.
The Non-Article 5 NATO Medal for NATO Logistical Support to the African Union Mission in Sudan
Non-Article 5 NATO Medal for NATO Logistical Support to the African Union Mission in Sudan obverse.
Origins: In 2005 NATO deployed troops in Sudan to assist the African Union with the logistics of its efforts to bring peace to the country.
Criteria: Thirty consecutive or sixty cumulative days of service in Sudan or in other African countries approved by NATO that host elements of the force providing direct support between July 13, 2005, and December 31, 2007.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 5 mm white stripe inset 5 mm from each edge bisected by a single thin silver thread.
Bars: The medal bears a plain bronze bar with AMIS in bas-relief.
Number: 66.
The Non-Article 5 NATO Medal for North Atlantic Council Approved NATO Operations and Activities in Relation to Africa
Origins: This medal was created to provide a uniform method of recognition for NATO operations conducted on the African continent, thus avoiding the need to create distinct medals for each operation as had been the case with the African Union Mission in Sudan.
Criteria: Thirty consecutive or sixty cumulative days of service under NATO command or control of all support activities in relation to the African Union approved by the North Atlantic Council, commencing September 1, 2008. There have been a number of separate missions conducted under the overall umbrella of activities approved by the North Atlantic Council: Staff Building Capacity within the African Union; strategic airlift support to the African Union Mission in Somalia (AMISOM); Operation Allied Provider, October 25 to December 12, 2008; Operation Allied Protector, March 24 to April 23, 2009, and May 1 to May 16, 2009; and Operation Ocean Shield, which commenced on August 17, 2009, and is ongoing.
Insignia: Standard NATO Service Medal.
Non-Article 5 NATO Medal for North Atlantic Council Approved NATO Operations and Activities in Relation to Africa obverse.
Ribbon: NATO blue, 36 mm wide, with a 5 mm white stripe inset 5 mm from each edge bisected by a single thin silver thread.
Bars: The medal bears a plain bronze bar with AFRICA in bas-relief.
Number: 355.
The Non-Article 5 NATO Medal for Service in nato Operation Unified Protector — Libya
Non-Article 5 NATO Medal for Service in Operation Unified Protector — Libya obverse.
Origins: In the spring of 2011 a multi-state coalition began a military intervention in Libya in response to events during the Libyan civil war. U.N. Security Council Resolution 1973 was implemented to create a Libyan no-fly zone and to take all necessary measures to prevent attacks on civilians by the government forces of Colonel Muammar Gaddafi.
Criteria: Thirty consecutive or sixty cumulative days of honourable service under NATO Operation Unified Protector (OUP), commencing on March 23, 2011, and ending on October 31, 2011. Service as part of or in direct support of OUP within the political boundaries, territorial waters, and airspace of Libya is eligible as well as service in or over the central southern Mediterranean Sea between ten degrees east and twenty-eight degrees east and south of forty-one degrees north and service at the support bases of Trapani and Sigonella and at NATO and Canadian Supporting Headquarters in Naples and Poggio Renatico, Italy.
Insignia: Standard NATO Service Medal.
Ribbon: NATO blue, 36 mm wide, with a 5 mm white stripe inset 5 mm from each edge bisected by a single thin silver thread.
Bars: The medal bears a plain bronze bar with OUP — LIBYA/LIBYE in bas-relief.
Number: 1,573.
NATO Medal certificate.
# 26
International Mission Medals
Since the end of the Second World War, Canada has participated in a wide variety of international organizations and commissions. These have ranged from entities such as the North Atlantic Treaty Organization and the United Nations to the Commonwealth, the European Union, and the Francophonie. With this increased international involvement, Canada's military has frequently been called upon to help in monitoring, truce-making, peacekeeping, and peacemaking missions.
The international mission medals that have been awarded to Canadians and recognized by the Canadian government are similar to the various U.N. and NATO medals. With the exception of the International Force East Timor Medal, all are the product of an international or regional organization and not a specific country. The International Force East Timor Medal is an Australian medal allowed by the Australian government to be awarded to non-Australian forces involved in the International Mission for East Timor (INTERFET). The most recent international mission medal comes from the European Union and is issued as part of the European Union Security Defence Policy Group, which is essentially the military operational wing of the European Union. The medal is sanctioned by the European Union and has been approved by the Canadian government with various bars.
Unfortunately, the international mission medals have tended to be of marginal quality and are aesthically drab. The 1973 International Commission for Supervision and Control in Vietnam Medal became affectionately known as the "Cracker Jack" or "tin pot" medal because of its shiny gold appearance and light weight.
As with the quality of these medals, there is no consistency with regard to the issue of award certificates. While certificates were issued with the 1973 ICCS Medal, they have generally not been issued with other international mission medals. There are no postnominals associated with any of these honours.
Because of the review of overseas service recognition undertaken in response to the campaign in Afghanistan, which resulted in the creation of several new Canadian service medals, it has been decided that in the future only U.N. and NATO medals will be integrated into the Canadian Order of Precedence, and if no such medal is available, then Canada will issue its own recognition instead of incorporating additional international or commission medals that vary considerably in criteria, quality, and administration.
The International Commission for Supervision and Control Service (ICSC) Medal
International Commission for Supervision and Control Medal obverse.
Origins: After the end of the Second World War, instability in French Indochina (Vietnam, Laos, and Cambodia) grew into a full-fledged conflict as the Vietminh waged a war of independence against French forces. In July 1954 a conference was held in Geneva and accords were signed. Three separate commissions were established, one each for Laos, Cambodia, and Vietnam. The accords further divided Vietnam into two separate parts: North and South. The International Commission for Supervision and Control (ICSC) was charged with supervising the ceasefires and the withdrawal of French military personnel, facilitating the return of refugees to their homes, and patriating of civilian and military prisoners. The commission was quite literally intended to supervise and control the implementation of the Geneva accords.
At the direction of the commission's Indian delegation, a medal was authorized by the ICSC in February 1961, though approval from the Department of National Defence did not come until March 1962. Even then the department was still considering whether or not to institute a Canadian General Service Medal rather than allow Canadian personnel to accept medals from international organizations. The General Service Medal project was never approved; therefore the ICSC Medal was sanctioned.
International Commission for Supervision and Control Medal reverse.
Criteria: Ninety days of consecutive or cumulative service as a member of the ICSC between August 7, 1954, and January 28, 1973. Persons killed while serving with the commission were automatically awarded the medal.
Insignia: A circular antiqued bronze medal 34 mm in diameter bearing the crossed flags of Canada and Poland with the arms of India in the centre and the dove of peace at the point where the flags cross circumscribed by INTERNATIONAL COMMISSION FOR SUPERVISION AND CONTROL with PEACE at the base. The reverse bears a map of Indochina with the names of the three countries — Vietnam, Laos, and Cambodia.
Suspender: A floral motif connects the mounting suspender to the medal. The suspender through which the ribbon passes appears as shoots of bamboo.
Ribbon: A 32 mm wide ribbon of three equal stripes of dark green, white, and red. The green represents India, the white peace, and the red Canada and Poland.
Bars: None.
Naming: Impressed capitals, abbreviated rank, initials, and surname. The impression was then filled with white paint.
Other: This award was superseded by the ICCS Medal. The medal was made by Barton's Jewellers of Bangalore, India, and was issued in a black leatherette box that was delivered covered in brown paper and tied with a purple ribbon.
Number: 1,550.
The International Commission of Control and Supervision (ICCS), Vietnam
International Commission of Control and Supervision Vietnam Medal obverse.
International Commission of Control and Supervision Vietnam Medal reverse.
Origins: After the Paris Peace Conference, the U.S.-led war in South Vietnam came to an end and American forces began pulling out. The International Commission of Control and Supervision (ICCS) was charged with monitoring the ceasefire and facilitating the exchange of prisoners. This mission was the successor to the International Commission for Supervision and Control (ICSC). In the early 1990s, at the direction of Minister of External Affairs Joe Clark, this medal was issued to civilians involved in the mission, and the Type II of the medal was struck. Members of the Canadian Armed Forces who received the original medal could apply to exchange their awards for the Canadian issue.
Criteria: Ninety consecutive or cumulative days of service as a member of the ICCS between January 28, 1973, and April 30, 1975. Persons killed while serving with the ICCS were automatically awarded the medal.
Insignia: There were two issues:
* • Type I: A bright gold-coloured circular medal 36 mm in diameter bearing the national emblems of the four countries involved (Canada, Hungary, Poland, and Indonesia) circumscribed by INTERNATIONAL COMMISSION OF CONTROL AND SUPERVISION. The reverse bears a laurel wreath with SERVICE — VIETNAM — 27-1-1973 on three separate lines. The medal is 1.6 mm thick.
* • Type II: A similar design to Type I. However, the medal is 3.2 mm thick, has a solid suspender, and the reverse features an enlarged maple leaf above VIETNAM 27-1-73.
Suspender: Two types:
International Commission of Control and Supervision Vietnam Medal Canadian issue obverse.
International Commission of Control and Supervision Vietnam Medal Canadian issue reverse.
* • Type I: An eyelet mounted on top of the medal through which passes a rounded rectangular clip that attaches the medal to a straight mounting bar.
* • Type II: A solid, straight suspender bar and claw.
Ribbon: A 36 mm wide ribbon consisting of eight alternating 4 mm stripes of red and white with a central 4 mm stripe of green.
Bars: None.
Naming: None.
ICCS certificate.
Other: Type I was struck by Shinchong Trading Company of Saigon. Type II was struck by Rideau Ltée and was redesigned by Bruce Beatty.
Number: 384.
The Multinational Force and Observers (MFO) Medal
Multinational Force and Observers Medal obverse.
Multinational Force and Observers Medal reverse.
Origins: The multinational force in the Sinai Peninsula was created as part of the Camp David Accord of 1979, which established peace between Egypt and Israel. The force is funded by the United States, Egypt, Israel, Japan, and Germany. Canadians have been involved with the multinational force since 1986. The purpose of the mission is to "observe, verify, and report" on the status of the area.
Criteria: Awarded for six months of service with the MFO stationed in the Sinai Peninsula.
Insignia: The medal is bronze in colour and 34 mm in diameter. A stylized dove of peace holding an olive branch is in the centre. The medal is circumscribed with MULTINATIONAL FORCE & OBSERVERS. The reverse is plain and bears UNITED IN SERVICE FOR PEACE.
Suspender: A straight suspender bar is connected to the medal by a small ring that passes through an eyelet on top of the medal.
Ribbon: Orange in colour, 36 mm wide, with a central 10 mm white stripe flanked on either side by 3 mm of dark green.
Bars: Bronze numerals are added for additional tours.
Naming: None.
Other: Each medal was issued in a plastic box and was accompanied by a miniature medal.
Number: 1,105.
The European Community Monitor Mission — Yugoslavia (ECMMY) Medal
European Community Monitor Mission Medal — Yugoslavia obverse.
Origins: On June 25, 1991, Slovenia and Croatia declared independence from Yugoslavia. Six days later the Yugoslav National Army (JNA) marched into Slovenia. On July 8 the Brioni Agreement came into force, effectively ending hostilities in Slovenia. Part of the agreement called for the European Community to establish a monitor mission to observe and report on the withdrawal of the JNA. As the conflict in the former Yugoslavia spread, the monitor mission grew to include Croatia, Bosnia, and Herzegovina. The main task was to "monitor political, humanitarian, security, military, and economic developments in the countries of the Western Balkans and deliver valid, firsthand reporting to the EU Council of Ministers and to the participating states." In December 2000 the name of the mission (which is still active) was changed to the European Union Monitor Mission (EUMM), which consists of approximately 120 monitors spread throughout the Balkans.
Criteria: Awarded for ninety days of service (consecutive or cumulative) as part of ECMMY. The mission began on September 8, 1991, and is current.
European Community Monitor Mission Medal — Yugoslavia reverse.
Insignia: A circular oxidized silver-coloured medal 36 mm in diameter. In the centre is a map of Yugoslavia defaced with EC MONITOR MISSION. The map is surrounded by twelve European Community stars. The reverse bears a large dove of peace in flight holding an olive branch in its beak.
Suspender: An eyelet mounted on top of the medal through which passes a ring and the ribbon.
Ribbon: The ribbon is 36 mm wide and consists of eleven stripes of varying sizes. Either edge has 2 mm of blue, 3 mm of white, 3 mm of red, 3 mm of blue, and 2 mm of yellow. The central stripe, measuring 10 mm wide, is blue. A circular emblem of the obverse of the medal is worn on the undress ribbon.
Bars: None.
Naming: None.
Other: This medal cancelled the award of the JUGOSLAVIJA bar to the Special Service Medal. It was issued in a dark blue leatherette box impressed with the European Union emblem in yellow and was accompanied by an undress ribbon and a lapel pin.
Number: 78.
The International Force for East Timor (INTERFET) Medal
Origins: In August 1999 the people of East Timor — a former Portuguese colony that had been invaded by Indonesia in 1975 — voted to become independent from Indonesia. Following the U.N.–monitored vote, Indonesian militia groups went on a rampage and drove more than two hundred thousand people from their homes. On September 15, 1999, the U.N. Security Council passed a resolution approving the creation of the International Force for East Timor (INTERFET). With the permission of the Indonesian government, INTERFET was authorized to restore peace and security in East Timor, to protect and support the United Nations Mission in East Timor (UNAMET) in its tasks, and to facilitate humanitarian aid efforts. The force was led by Australia, and sixteen other nations, including Canada, participated. The mission lasted from October 1999 until April 2000.
International Force for East Timor Medal obverse.
International Force for East Timor Medal reverse.
Criteria: The Canadian criteria for the INTERFET Medal is a minimum of thirty days of cumulative service in the area of operations — defined as East Timor, the sea adjacent to East Timor to a distance of twelve nautical miles, or the staging base at Darwin, Australia.
Insignia: An oxidized silver-coloured medal 38 mm in diameter, the obverse bearing a dove holding an olive branch within its beak overlaying a map of East Timor and circumscribed by INTERNATIONAL FORCE EAST TIMOR. There is also an olive branch at the base of the medal. The reverse is plain and circumscribed with TOGETHER AS ONE FOR PEACE IN EAST TIMOR. The highlights of the medal are polished.
Suspender: A footed suspender with an Australian Federation star in the centre mounted directly onto the top edge of the medal.
Ribbon: A 32 mm wide ribbon with a 2 mm central red stripe flanked on either side by 4 mm of green, 7 mm of white, and 4 mm of blue.
Bars: None.
Naming: None.
Other: Although the medal was officially issued to Canadians unnamed, the reverse is plain to allow for engraving. This medal is quite unusual in that it is an Australian award that was awarded to Canadians. The Queen — as Queen of Australia — approved the creation of the International Force for East Timor Medal on March 25, 2000, at Government House, Canberra.
Number: 760.
The Common Security and Defence Policy Service Medals
European Security and Defence Policy Service Medal obverse.
Since the early 1990s, the European Union has taken an increasingly active role in various international missions both within Europe and beyond. These actions have been taken within the framework of the European Security and Defence Policy (ESDP), which is administered by the Council of the European Union. To recognize those personnel participating in ESDP missions, the European Union authorized the striking of a standard ESDP medal to which various service bars are attached. To date, eighteen bars have been created for this medal; seven have been awarded to Canadians as listed below.
Criteria: Awarded for thirty days of continuous or accumulated service as part of a specific European Union mission:
* • EUPM: The European Union Police Mission in Bosnia Herzegovina commenced on January 1, 2003, and followed the U.N. International Police Task Force that had been active in Bosnia to maintain the rule of law and assist with the fostering of peace and order. Approximately five hundred police officers from thirty-two countries made up the mission.
* • ARTEMIS: In the summer of 2003, increasing unrest in the Democratic Republic of the Congo and an impending humanitarian disaster prompted the European Union to assist the United Nations Organization Mission in the Democratic Republic of the Congo (MONUC) to evacuate people and provide supplies to local residents. The mission began on May 30, 2003, and concluded on September 1, 2003. The Canadian name for the operation was Operation Caravan.
* • CONCORDIA: From 1992 until 1995 the United Nations Preventative Deployment Force was charged with ensuring the integrity of the Former Yugoslav Republic of Macedonia's borders, a responsibility that was assumed by NATO in 1995. Concordia began in August 2001 and lasted until the end of September 2001, though NATO forces remained in Macedonia until March 31, 2003. Under the authority of the U.N. Security Council, the European Union launched Concordia. The purpose of this operation was to provide a stable and secure environment and to allow for the implementation of the August 2001 Ohrid Framework Agreement, which set out to promote democracy, foster better relations between the various ethnic groups living in the Former Yugoslav Republic of Macedonia, and improve relations between Macedonia and the European Union. More than four hundred military personnel from twenty-six countries participated in the mission. Awarded for service between March 31 and December 15, 2003.
European Security and Defence Policy Service Medal ribbon for headquarters service.
* • ALTHEA: As the number of NATO troops involved in the Stabilization Force in Bosnia (SFOR) was reduced in December 2004, the European Union Force (EUFOR) began taking on more responsibility. The long-term goals of this operation are to ensure a peaceful and multi-ethnic Bosnia Herzegovina, to foster peaceful co-operation with its neighbours, and to encourage eventual European Union membership. The Canadian name for this mission is Operation Boreas. Awarded for service after December 2, 2004.
* • EUPOL-AFG: The European Union Police Mission in Afghanistan is a police-training operation headed by the European Union with the goal of establishing sustainable and effective policing arrangements in all parts of Afghanistan. Awarded for service after September 1, 2007; the mission is due to conclude on December 31, 2016.
European Security and Defence Policy Service Medal reverse.
* • EUPOL COPPS: The European Union Coordination Office for Palestinian Police Support provides training support for building the capacity of the Palestinian territories in the areas of policing and criminal justice. Awarded for services commencing after August 29, 2008.
* • EULEX KOSOVO: After the independence of Kosovo and under the umbrella of the U.N. Interim Administrative Mission in Kosovo, the European Union Rule of Law Mission in Kosovo includes police officers, prosecutors, and judges. Awarded for services commencing after May 16, 2009.
Insignia: The standard ESDP medal is an oxidized silver-coloured medal 35 mm in diameter. The obverse bears twelve five-pointed stars around the circumference of the medal. These stars symbolize the unity of Europe. The reverse is plain except for PRO PACE UNUM ("United for peace") on three lines in the centre. Each medal is awarded with a mission-specific bar.
Suspender: A ball is mounted on top of the medal through which passes a ring and the ribbon.
Bars: Bars are awarded with every medal, and a small version of the bar is worn on the undress ribbon. Silver numerals can be worn to signify additional tours. Each bar is 36 mm wide and 6 mm tall and bears mission-specific text (see the list of missions above in "Criteria"; the capit-alized mission names appear on the bars).
Ribbon: A 36 mm wide ribbon with three equal stripes of yellow and European Union blue.
EUPM Bar.
Artemis Bar.
Concordia Bar.
Althea Bar.
Naming: None.
Other: Originally known as the "European Security and Defence Policy Service Medals," this award is manu-factured by Eekellers-Centini of Belgium. A letter E is impressed on the reverse mounting ball on top of the medal. The medals are issued in a rectangular blue plastic case. The lid is marked with a circle of twelve stars and the name of the mission (i.e., ARTEMIS) and ESDP SERVICE MEDAL.
Number:
* • EUPM: 6.
* • ARTEMIS: 53.
* • CONCORDIA: 1.
* • ALTHEA: 70.
* • EUPOL-AFG: 12.
* • EUPOL COPPS: Number not available.
* • EULEX KOSOVO: 1.
# 27
Special Medals
This chapter examines three medals that do not easily fit into the broader Canadian honours system. While the Queen's Medal for Champion Shot, an award for marksmanship, has been a constant part of the honours system for nearly a century, the recently established Polar Medal and Sovereign's Medal for Volunteers are very new creations for which there was no precedent in the post-1967 system. These last two honours were established in part as a result of the 2011 honours review.
The Queen's Medal for Champion Shot of the Canadian Forces
Proficient marksmanship within the armed forces is a talent that continues to be universally recognized. In many countries, marksmanship is recognized through the award of a badge worn on the uniform. Since the time of New France, it has been a tradition for the Crown to organize shooting competitions, both for the regular and reserve forces. The history and purpose of the Queen's Medal for Champion Shot is almost identical to that of its antecedent, which is examined in chapter 9. This medal was identical in appearance to the previously issued Queen's Medal for Champion Shot of the Army; the difference was that the competition was now open to members of all branches of the Canadian Armed Forces.
On July 10, 1991, the current Queen's Medal for Champion Shot was established. This decision was in part made to incorporate the Queen's Canadian titles and CANADA into the obverse of the medal, thereby differentiating it from the British issue. The medal, its administration, and the Queen's Medal Competition remain essentially the same as the pre-1991 Queen's Medal for Champion Shot of the Military Forces. The idea of creating a separate Canadian award was in part prompted by the Australian government's establishment of a separate Australian Medal for Champion Shot in 1988.
Queen's Medal for Champion Shot of the Canadian Forces Medal obverse (first type suspender).
Queen's Medal for Champion Shot of the Canadian Forces Medal reverse.
Criteria: Awarded to the member of the Canadian Armed Forces or RCMP who obtains the highest aggregate score in stages one and two of the Queen's Medal Competition. Two medals are awarded yearly, one to a member of the Canadian Armed Forces (regular) and one to a member of either the Canadian Armed Forces (reserve) or the RCMP.
Insignia: A silver-coloured circular medal 36 mm in diameter. The obverse bears a crowned effigy of Queen Elizabeth II circumscribed by the legend ELIZABETH II DEI GRATIA REGINA with CANADA at the bottom. The reverse depicts the figure of Fame rising from her throne, facing left, with a horn in her left hand. With her right hand she is crowning with a laurel wreath a warrior facing right. The warrior's weight is on his right foot because his left foot is raised and resting on the dais to support on his left knee a target having three arrows in the centre. In his right hand he holds a bow and a quiver full of arrows.
The obverse was designed by Bruce Beatty, while the reverse — which is identical to the British issue instituted in 1869 — was designed by the noted English engraver J.B. Wyon. The name and rank of the recipient are engraved on the rim of the medal. Initially, there were some problems with the naming, since it distorted the rhodium plating on the medal. The first award was made in 1992.
Suspension: Two types of suspenders have been employed:
* • Type I: A non-swivelling claw-footed suspender with a straight bar.
* • Type II: A footed fleur-de-lys attached to a straight bar.
The style of suspender was changed in 2002 in order to reduce the cost of striking the medal.
Ribbon: A 32 mm wide ribbon with a stripe of red in the centre flanked on each side by 3 mm stripes of black, cream (white), and black.
Bars: Each medal is issued with a date bar indicating the year of the award. Until 2002 the date bars were riveted to the suspension of the medal, as had been done with earlier issues. However, these bars are now simply sewn onto the ribbon of the medal. If a recipient wins a later competition, he or she receives another bar bearing the year of the award.
Queen's Medal for Champion Shot obverse (second type suspender and bar).
Naming: Engraved capitals, abbreviated rank, initials, and surname.
Postnominals: None.
Other: The medal is presented in a black or blue leatherette case with a vice-regal lion embossed in gold on the lid.
Number: 31 medals, multiple bars.
The Polar Medal
Polar Medal obverse.
Polar Medal reverse.
The Polar Medal was created to recognize scientific, geographic, and exploratory work undertaken in Canada's polar region and also potentially the Antarctic. Up to fifty medals will be awarded annually, and this honour in part replaces the Governor General's Northern Medal, which was established by Adrienne Clarkson. The inaugural investiture took place on July 8, 2015, in Whitehorse, Yukon Territory, where ten medals were presented.
The overall design of the medal is based on the Polar Medal 1818–55, which was awarded to those involved in the Franklin Expedition and subsequent recovery-and-rescue missions associated with it. The United Kingdom, Australia, and New Zealand all possess similar honours, though the Australia and New Zealand medals are only awarded for service in the Antarctic. The design of each medal links back to the original Polar Medal, being octagonal in shape and hung from a white ribbon, though the ribbon for the Australia Antarctic Service Medal is edged in blue.
Criteria: Service in an expedition or undertaking approved by the Government of Canada in the polar regions — either Arctic or Antarctic.
Insignia: A cupro-nickle octagonal medal 34 mm in diameter. The obverse bears an effigy of the Queen wearing the Canadian Diadem and a fur collar circumscribed by the legend ELIZABETH II DEI GRATIA REGINA with CANADA at the bottom. The reverse depicts the RCMP vessel St. Roch in the foreground with ice mountains in the background. The reverse was designed by Major Carl Gauthier.
Suspension: A pair of curved narwhal tusks joined in the centre by a Northern Star and a straight bar from which the ribbon hangs.
Ribbon: White moiré 38 mm in width.
Bars: A silver bar with raised edges and a centred maple leaf may be awarded for subsequent expeditions.
Naming: Engraved capitals around the rim.
Postnominals: None.
Number: 10.
The Sovereign's Medal for Volunteers
Sovereign's Medal for Volunteers obverse.
Originally established as a Governor General's Award by Governor General Romeo LeBlanc, the Sovereign's Medal for Volunteers began as the Caring Canadian Award in 1995. The criteria was left broad to encompass a wide range of contributions, notably LeBlanc desired that the award be given to those who have made significant, sustained, unpaid contributions to their communities in Canada or abroad.
Recipients of the Caring Canadian Award were presented with a certificate signed by the governor general along with a gold lapel pin bearing the logo of the award. From its creation until 2015 more than 1,700 awards were bestowed by the governor general.
In 2015 the Sovereign's Medal for Volunteers was established by Queen Elizabeth II, and all recipients of the Caring Canadian Award were automatically awarded the new medal — the award having been converted into a national honour. Up to one thousand medals will be awarded annually.
Sovereign's Medal for Volunteers reverse.
Criteria: Sustained, unpaid community service to a community in Canada or abroad. Service must have been rendered over an extended period of time.
Insignia: A circular medal, silver in colour, 36 mm in diameter, the obverse bearing a contemporary effigy of Queen Elizabeth II wearing a Canadian Diadem made of maple leaves and snowflakes circumscribed by the legend ELIZABETH II DEI GRATIA REGINA with CANADA at the bottom. The reverse displays a representation of a large and a small heart interlaced, set with small maple leaves, and surmounted by a coronet, all within a sunburst pattern. The reverse design was inspired by the Scottish Luckenbrooth brooches traded in North America in the seventeenth century. The reverse concept was devised by Darcy DeMarsico of the Chancellery of Honours.
Suspension: A claw-footed suspender with a circular ring through which the ribbon passes.
Ribbon: A 32 mm wide ribbon maroon in colour with five alternating 1 mm stripes of gold and four of blue.
Bars: None.
Naming: Engraved capitals around the rim.
Postnominals: None.
Number: 1,733.
# 28
Commemorative Medals
The practice established in 1887 with the striking of a wearable medal to commemorate the Sovereign's jubilee is one that continues to be observed in Canada. From Queen Victoria's Golden Jubilee through to the coronation of Queen Elizabeth II, Canadians were recognized with a variety of imperial commemorative medals. A full examination of the British tradition of awarding commemorative medals for wear is examined in chapter 8.
With the Queen Elizabeth II Silver Jubilee Medal, the 125th Anniversary of Confederation Medal, the Queen Elizabeth II Golden Jubilee Medal, and most recently the Queen Elizabeth II Diamond Jubilee Medal, the practice of striking and awarding commemorative medals to a broad range of Canadian citizens has become a firmly entrenched element of the Canadian honours system. Indeed, it is through these medals that the broadest cross-section of Canadian society has been recognized. That these awards continue to serve as Royal Favours and are predominantly used to reward citizens who have made significant contributions at the local level is a testament to how "Canadianized" their awarding has become.
The first Canadian commemorative medal intended for wear was struck in 1967 to mark the Centennial of Confederation. In 1917 there had been muted discussion about striking a Golden Jubilee of Confederation Medal, though this idea was not well received in light of the Great War. In 1967 there was no such impediment, and in many ways the Centennial Medal was the catalyst for the creation of a broader Canadian honours system.
The idea for the Centennial Medal came from Undersecretary of State Jean Miquelon in 1963. Initially, the Department of National Defence wanted the medal to be awarded only to members of the armed forces. This idea was discarded, and it was decided that the as-yet-unnamed medal was to be open to all Canadians. On January 22, 1964, Paul Hellyer, the new minister of national defence, gave his approval for the award of the medal to both military personnel and civilians.[1] Secretary of State Maurice Lamontagne received a letter from the Centennial commissioner in March 1964 outlining the commission's view that the issue of creating a commemorative medal was for the government and not the Centennial Commission, to consider.[2]
Hellyer and Lamontagne presented a joint memorandum to the Cabinet on May 11, 1965, in which they outlined the insufficiency of the existing honours policy as embodied in Cabinet Directive 30 of 1956. The conclusion outlined that the Centennial Medal would appropriately recognize "the specially valuable service of many Canadian citizens."[3] The submission was discussed at the May 20 Cabinet meeting, though Prime Minister Lester Pearson pushed the project aside: "The prime minister indicated at the May 20 meeting] he wished to discuss the proposal further with Mr. Lamontagne. Mr. Pearson was inclined to the view that the institution of the Centennial Medal should be linked to the establishment of the Canada Medal."[[4] In other words, Pearson was more concerned with getting the nascent Order of Canada project off the ground than establishing a single medal.
Approval for the Centennial Medal was given at the same time as the Cabinet approved the creation of the Order of Canada. The Commonwealth tradition of awarding commemorative medals was thus firmly established in Canada, and it is a tradition that continues to this day. Every Canadian commemorative and jubilee medal has been issued with a distinctive certificate. There are no postnominals associated with any of these commemorative medals.
The Canadian Centennial Medal (1967)
Canadian Centennial Medal obverse.
Canadian Centennial Medal reverse.
Canadian Centennial Medal Birks issue reverse.
Origins: The Centennial Medal finds its origin in the 1947 Independence Medals issued by India and Pakistan. Both of these countries inaugurated the tradition of striking commemorative medals on important national occasions not directly associated with the Sovereign.
Criteria: Awarded to Canadian citizens who were deemed to have made a significant contribution to their community, province, or the country as a whole.
Canadian Centennial Medal certificate.
Insignia: A circular silver medal 36 mm in diameter. The obverse bears a maple leaf superimposed with a Royal Cypher circumscribed with CONFEDERATION CANADA CONFÉDÉRATION. The reverse depicts the Royal Arms of Canada with the dates 1867–1967 in the base. The medal was struck from sterling silver by the Royal Canadian Mint. The design concept for the medal and ribbon was developed by Harold Diceman, and the formal design was done by Bruce Beatty.
Suspension: Claw suspender with a graduated suspender bar.
Ribbon: Red and white, 36 mm wide, with 5 mm of red on either edge and 4 equally spaced 1 mm stripes of red set against a white background.
Naming: Issued unnamed, though those personally presented by the governor general and those awarded to Government House staff members were officially impressed in block capitals with the person's full given names and surname.
Other: The medal was presented in a Farrington box (made of metal), its lid stamped with the Royal Arms of Canada in gold (the same case in which the Canadian Forces' Decoration was awarded). The medal was also accompanied by two undress ribbon bars. Several gold-plated specimens exist that were originally to be awarded to the Expo 67 commissioners, though there is no evidence that the awards were ever presented.
The medal was issued with a certificate printed on white bond paper with the Royal Arms of Canada at the top in red and the name of the recipient typed below. Certificates were issued in either English or French.
In the mid-1990s the Chancellery at Rideau Hall in Ottawa ran out of the original issue of the Centennial Medal struck by the Royal Canadian Mint. The contract to strike replacement issues was awarded to Henry Birks & Sons, and they were made by Pressed Metal Products. The Birks issues, though struck in sterling silver, are distinguishable from the mint's original striking because they are slightly thinner and the upper bar on the reverse of the suspender is marked in raised text with BIRKS STERLING.
Number: 29,500.
The Commemorative Medal for Her Majesty Queen Elizabeth II's Silver Jubilee (1977)
Queen Elizabeth II's Silver Jubilee Medal obverse.
Queen Elizabeth II's Silver Jubilee Medal reverse.
Queen Elizabeth II's Silver Jubilee Medal certificate.
Origins: Created as part of Queen Elizabeth II's Silver Jubilee program in Canada, this medal followed the tradition set when a special medal was struck for King George V's Silver Jubilee in 1935. The program was officially established on February 6, 1977.
Criteria: Awarded to Canadian citizens who were deemed to have made a significant contribution to their community, province, or the country as a whole. Recipients were selected by a variety of government officials, including elected politicians.
Insignia: A circular silver medal 32 mm in diameter. The obverse bears a crowned effigy of Queen Elizabeth circumscribed by ELIZABETH II DEI GRATIA REGINA FID DEF. The reverse bears a stylized maple leaf with CANADA above and the dates 1952–1977 below separated by the Royal Cypher. The medal was struck from sterling silver by the Royal Canadian Mint. While the obverse was designed by the Royal Mint staff in London and is identical to the British issue, the reverse was designed by the Canadian artist Dora de Pédry-Hunt.
Suspension: Claw-footed suspender with a ring 12 mm in diameter.
Ribbon: A 32 mm wide watered white ribbon with a central 1 mm red stripe bordered on either side with 2 mm blue stripes and edged with 1 mm stripes of red. This ribbon is the same as used for the British issue.
Naming: The award was issued unnamed, though those personally presented by the governor general and those awarded to Rideau Hall staff members and certain VIPs were officially impressed in block capitals with the person's full given names and surname.
Other: The medal was presented in a small red cardboard box with a flip lid. The lid was impressed with the Royal Cypher in silver along with the dates 1952–1977. A grey-brown-coloured certificate was issued with each medal. The 1977 Jubilee emblem was embossed in silver at the top of the certificate, and the name of the recipient was typed in the middle.
Number: 30,000.
The Commemorative Medal for the 125th Anniversary of Confederation (1992)
Commemorative Medal for the 125th Anniversary of Confederation obverse.
Commemorative Medal for the 125th Anniversary of Confederation reverse.
Origins: Part of the 125th Anniversary of Confederation program — "Canada 125" — this medal is based on the 1967 Centennial Medal. It was formally established on May 7, 1992.
Criteria: The medal was awarded to people who made a significant contribution to the well-being of their fellow citizens, their community, or to Canada. Recipients were selected by a variety of government officials, including elected politicians.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse bears the Royal Cypher surmounted by the Royal Crown superimposed on a single maple leaf circumscribed by CONFEDERATION — CONFÉDÉRATION with the dates 1867–1992 in the base. The reverse bears the shield of the Royal Arms of Canada encircled by the motto of the Order of Canada, surmounted by the vice-regal lion, and circumscribed with the national motto A MARI USQUE AD MARE at the bottom. The medal is made of a copper-and-zinc alloy, rhodium-plated, issued unnamed, and was manufactured by Rideau Ltée and Guthrie Woods Products Ltd. This medal was designed by Bruce Beatty and is a revised version of the 1967 Centennial Medal.
Suspension: A footed suspender bearing a maple leaf in the centre mounted directly onto the top edge of the medal.
Ribbon: A 32 mm wide white ribbon edged with 4.5 mm stripes of blue with 5 1 mm red stripes equally spaced between the two blue stripes.
Naming: None.
Commemorative Medal for the 125th Anniversary of Confederation certificate.
Other: The medal was issued in a blue cardboard box with the Royal Arms of Canada and the dates 1867–1992 stamped on the lid. Medals were awarded with a certificate bearing the recipient's name with the Royal Arms of Canada printed in silver at the top.
Number: It was projected that fifty thousand would be presented, but only forty-two thousand were issued due to problems with the program's administration and a lack of nominations.
The Commemorative Medal for Her Majesty Queen Elizabeth II's Golden Jubilee (2002)
Origins: Part of Queen Elizabeth II's Golden Jubilee program, the medal was formally established on February 15, 2002.
Criteria: Awarded to Canadian citizens who have made a significant contribution to Canada or to a particular province, territory, region, or community within Canada, or who have made an outstanding achievement abroad that brings credit to Canada. Recipients were selected by a variety of government officials, including politicians.
Queen Elizabeth II's Golden Jubilee Medal obverse.
Queen Elizabeth II's Golden Jubilee Medal reverse.
Insignia: A circular gold-coloured medal 32 mm in diameter. The obverse bears an effigy of Queen Elizabeth II wearing the George IV Diadem circumscribed by the legend REINE DU CANADA — QUEEN OF CANADA. The reverse bears the Royal Crown above a single maple leaf upon which is superimposed the Royal Cypher, all of which is circumscribed by 1952 — CANADA — 2002. The medal was struck by the Royal Canadian Mint and made from gold-plated bronze. It was designed by a committee at the Chancellery.
Queen Elizabeth II's Golden Jubilee Medal certificate.
Suspension: An eyelet screwed into the top of the medal through which passes a ring and the ribbon.
Ribbon: A 32 mm wide ribbon with a central 2 mm red stripe bordered on each side by 3 mm of white, 10 mm of blue, and 2 mm of red at the edge. This ribbon is the same as used for the British issue.
Naming: A few medals issued to Rideau Hall staff are found named in impressed block capitals.
Other: The medal was issued in a blue cardboard box with the Canadian Golden Jubilee emblem (the Royal Cypher EIIR with a flourish of maple leaves with the numeral 50) stamped in gold on the lid. Awards were accompanied by a white certificate bearing the Golden Jubilee emblem and the fiftieth anniversary of Canadian governors general logo at the bottom with the recipient's name laser-printed in the centre.
Number: 46,470.
The Queen Elizabeth II's Diamond Jubilee Medal (2012)
Queen Elizabeth II's Diamond Jubilee Medal obverse.
Origins: Part of Queen Elizabeth II's Diamond Jubilee program. The establishment of the medal and program was announced on February 3, 2011, more than a year before the beginning of the jubilee year, but the medal was officially created on January 13, 2012.
Criteria: Awarded to Canadian citizens who have made a significant contribution to Canada or to a particular province, territory, region, or community within Canada, or who have made an outstanding achievement abroad that brings credit to Canada. Recipients were selected by a variety of government officials, including elected politicians.
Insignia: Struck in nickel silver, the 32 mm medal is silver in colour and displays on the obverse a crowned effigy of the Queen circumscribed by her Canadian titles, ELIZABETH II DEI GRATIA REGINA — CANADA. The reverse bears a diamond with the Royal Cypher in the centre, and protruding from each side of the diamond is a stylized maple leaf, the field being made up of a pattern of many small diamonds. The dates 1952 and 2012 also appear on the reverse along with the Latin motto VIVAT REGINA ("Long live the Queen"). The medal itself was designed by Fraser Herald Cathy Bursey-Sabourin of the Canadian Heraldic Authority.
Suspension: A claw-footed suspender with a ring 15 mm in diameter.
Queen Elizabeth II's Diamond Jubilee Medal reverse.
Queen Elizabeth II's Diamond Jubilee Medal certificate.
Ribbon: The ribbon of the medal is yet another permutation of the 1953 Coronation Medal ribbon: a dark red ribbon 32 mm in width edged with 2 mm of blue with a 2 mm central stripe of dark red bordered on each side by a 3 mm stripe of white. Major Carl Gauthier devised this new version of the ribbon based on the logic used for previous ribbons.
Naming: Issued unnamed. The sixty medals presented at the inaugural investiture held on February 6, 2012, and the medals issued to some senior dignitaries were engraved with the date 2012-II-06 on the rim.
Other: The medal was issued in a maroon cardboard box with the Canadian Diamond Jubilee emblem stamped in silver on the lid. Awards were accompanied by a white certificate bearing the Diamond Jubilee emblem in colour with the recipient's name laser-printed in the centre and the governor general's signature at the bottom. The governor general signed all sixty thousand certificates by hand.
Number: 60,000.
# 29
Royal Canadian Mounted Police Long Service Medal
The Royal Canadian Mounted Police Long Service Medal is the oldest continuously awarded long service medal in the Canadian honours system and one of the oldest in the Commonwealth. In the summer of 1925, Commissioner Cortlandt Starnes began to investigate how a special long service medal could be established. By September 22 of the same year, he had submitted a memorandum to the minister of justice suggesting the creation of a Royal Canadian Mounted Police Officers' Decoration and a Royal Canadian Mounted Police Long Service Medal.
The project seems to have received little attention. It was not until September 30, 1927, that Starnes submitted a secondary report that altered the proposed length of service from twenty-five to twenty years. Prime Minister William Lyon Mackenzie King gave his consent for the project as did Governor General Lord Willingdon, and on December 5, 1928, an order-in-council was issued creating the medal and decoration.
The decoration was to be awarded to officers who had fulfilled twenty-five years of commissioned service, while the medal was to be awarded to officers, non-commissioned officers, and constables upon completion of twenty-five years of service. It is likely that the decoration was meant to look something like the Colonial Auxiliary Forces Officers' Decoration.
Despite the adoption of an order-in-council, no further action was taken. The idea for a medal was revived in 1932 by Major-General Sir James Howden MacBrien, the commissioner of the RCMP. A long-serving Canadian army officer, MacBrien had seen service in the South African War (1899–1902) and in the First World War. Shortly after being appointed commissioner, he broached the topic with Prime Minister R.B. Bennett. MacBrien felt that the RCMP should have a long service medal similar to that awarded to members of the Canadian militia, but unlike Starnes he did not see a need for a separate officers' decoration. Bennett — who, it will be remembered, resumed the practice of submitting civilian honours lists — was a keen supporter of the proposal. On January 14, 1933, the old order-in-council that had created the Officers' Decoration and Long Service Medal was cancelled, and a new one ordered that a Royal Canadian Mounted Police Long Service Medal be issued.
RCMP Long Service Medal George V obverse.
RCMP Long Service Medal Type I reverse.
The medal was formally established by King George V on March 6, 1934. Already elated that the Canadian government was resuming civilian honours lists, the King was enthusiastic about the new medal. It was the first uniquely Canadian honour established by the King in Right of Canada.
Formal announcement of the medal's establishment came at Standish Hall in Hull, Quebec, on April 26, 1934, as Commissioner MacBrien informed the gathering that His Majesty the King had been pleased to approve the establishment of the medal. The first investiture took place in the Drill Hall at Cartier Square on March 12, 1935. Governor General Lord Bessborough invested seventy-one members and former members of the force.
Other parts of the British Empire, such as Newfoundland, South Africa, Malta, and Cyprus, had already established long service medals for their police forces. The Canadian request was not unusual, especially since the RCMP was one of the most respected police forces in the empire. A few members of the Royal North West Mounted Police had been awarded the Imperial Service Medal, but until 1935 there was no medal unique to the Canadian police.
In 1954 an order-in-council was adopted to establish a series of clasps in bronze, silver, and gold. These are awarded for twenty-five, thirty, and thirty-five years of service. In 2004 a gold-and-silver bar with four stars was approved to recognize forty years of service. In 2012 a gold-and-silver bar with five stars was established to recognize forty-five years of service.
The medal was designed by the staff at the Royal Canadian Mint and continues to be struck there. It is the only long service medal that continues to be struck in silver. Each medal is awarded with a certificate bearing the recipient's name.
The Royal Canadian Mounted Police Long Service Medal
RCMP Long Service Medal George VI
Type I obverse.
Origins: Created as part of the broader trend throughout the Commonwealth in the 1920s and 1930s of dominion and colonial governments establishing police long service medals. The criteria and design are similar to the Colonial Auxiliary Forces Long Service Medal, which was awarded to members of the Canadian Militia from 1901 until 1931.
Criteria: Awarded to any officer, non-commissioned officer, or constable who bears an irreproachable character and has completed not less than twenty years of service.
Insignia: The medal has borne the effigies of Canada's Sovereigns since King George V, and there have been four obverse types: King George V, King George VI Type I, King George VI Type II, and Queen Elizabeth II. There have been three reverse types: the first displaying the RCMP badge with a Tudor Crown, the second displaying the RCMP badge with a Royal Crown and English legend, and the third displaying the RCMP badge with a Royal Crown and French legend. The medal is circular and silver and is 36 mm in diameter. There have been four different obverse designs:
* • George V crowned in robes: GEORGIVS V REX ET IND : IMP :.
* • George VI 1st type: GEORGIVS VI REX ET IND. IMP.
* • George VI 2nd type: GEORGIVS VI DEI GRATIA REX.
* • Elizabeth II uncrowned: Elizabeth Dei Gratia Regina.
RCMP Long Service Medal George VI Type II obverse.
RCMP Long Service Medal Elizabeth II obverse.
There have been three reverses:
RCMP Long Service Medal Type II reverse in English.
RCMP Long Service Medal Type II reverse in French.
* • RCMP badge with a Tudor Crown (for the George V, George VI issues, and early Elizabeth II).
* • RCMP badge bearing St. Edward's Crown and For Long Service and Good Conduct.
* • RCMP badge bearing St. Edward's Crown and Pour ancienneté et bonne conduite.
Suspender: Two types of suspenders have been issued with this medal:
* • Type I swivelling claw suspender with a straight bar.
* • Type II non-swivelling claw-footed suspender with a straight bar.
Bronze clasp.
Silver clasp.
Gold clasp.
Gold and silver clasp with four stars.
Gold and silver clasp with five stars.
Ribbon: Blue, 32 mm wide, with a 3 mm yellow stripe set 6 mm from either edge.
Clasps: There are five types of clasps. Only the clasp for the longest service is worn on the ribbon:
Early RCMP Long Service Medal certificate, 1959.
RCMP Long Service Medal certificate, 1979.
* • Bronze clasp with a single star in the centre awarded for twenty-five years of service.
* • Silver clasp with two stars equally spaced in the centre awarded for thirty years of service.
* • Gold clasp with three stars equally spaced awarded for thirty-five years of service.
* • Gold and silver clasp with four stars equally spaced awarded for forty years of service.
* • Gold and silver clasp with five stars equally spaced awarded for forty-five years of service.
RCMP Long Service Medal certificate, 1981.
RCMP Long Service Medal certificate, 2005.
Naming: Impressed capitals with initials and surname.
Postnominals: None.
Other: From 1934 until 1948 the medal was made of sterling silver (.925 silver). From 1948 until 1975 the medal was struck in .800 silver, which is more durable. Since 1975 the medal has once again been struck in sterling silver. All medals are named in block capitals. They were originally housed in a plain white rectangular cardboard box, but since the 1960s most have been issued in a square blue leatherette case with a miniature version of the RCMP badge on the lid.
NUMBER OF RCMP LONG SERVICE MEDALS PRESENTED
Type | Obverse Effigy | Reverse | Suspender | Silver | Dates Issued | Number Issued
---|---|---|---|---|---|---
Type I | George V | 1920 Badge | Claw | 925 | 1935–37 | 388 (392 struck)
Type II | George VI
(Ind Imp) | 1920 Badge | Claw | 925 | 1938–48 | 523
Type IIIa | George VI
(no Ind Imp) | 1920 Badge | Claw | 925 | 1948 | 79
Type IIIb | George VI
(no Ind Imp) | 1920 Badge | Claw-footed | 800 | 1949–50 | 204
Type IIIc | George VI
(no Ind Imp) | Re-engraved
1920 Badge | Claw-footed | 800 | 1951–52 | 46
Type IVa | Elizabeth II | Re-engraved 1920 Badge | Claw-footed | 800 | 1953–55 | 482
Type IVb | Elizabeth II | English
1954 Badge | Claw-footed | 800 | 1955–67 | 1,329
Type IVc | Elizabeth II | English
1954 Badge | Claw-footed | 925 | 1967–95 | 21,000
Type IVd | Elizabeth II | French
1954 Badge | Claw-footed | 925 | 1978–98
Type IVe | Elizabeth II
Brushed Finish | English
1954 Badge | Claw-footed | 925 | 1998–present
Type IVf | Elizabeth II
Brushed Finish | French
1954 Badge | Claw-footed | 925 | 1998–present
# 30
Canadian Forces' Decoration
The Canadian Forces' Decoration is one of the most widely recognized awards in the Canadian honours system. In its modern form the tradition of awarding long service medals was initiated with the creation of the Royal Navy Long Service and Good Conduct Medal in 1830. This medal was originally awarded for twenty-one years of service in the ranks of the Royal Navy. Later, the British Army instituted a similar medal, and throughout the late Victorian era and after the First World War a wide variety of long service medals and decorations was created by the British government for use throughout the British Empire. The Canadian Armed Forces used these awards until just after the Second World War when the Canadian Forces' Decoration (CD) was established.
Prior to the CD, Canadian service personnel had been eligible for any one of ten separate long service awards, depending on which service they were enrolled in and what rank they held. There was the Royal Canadian Navy Long Service Medal, the Royal Canadian Navy Voluntary Reserve (RCNVR) Long Service Medal, the Royal Canadian Navy Reserve (RCNR) Long Service Medal, the RCNVR Officers' Decoration, and the RCNR Officers' Decoration for members of the navy. The army had the Army Long Service and Good Conduct Medal, the Efficiency Medal, and the Efficiency Decoration. Those in the air force were eligible for the Royal Canadian Air Force Long Service and Good Conduct Medal or the Air Efficiency Award (see chapter 7).
Brooke Claxton, the Canadian minister of defence from 1946 until 1954, had a keen interest in honours and awards and was the central supporter of plans for a unique Canadian long service award. Clerical staff at the Department of National Defence had long complained that administering the various separate long service awards was time-consuming and complex, and plans for a separate Canadian long service award emerged from the department in 1947.
The Cabinet approved the establishment of the Canadian Forces' Decoration on December 15, 1949, and a submission was forwarded to King George VI. The King approved the creation of the CD on March 8, 1950. After 1949 the CD superseded all of the above-listed awards.
Canadian Forces' Decoration George VI issue in silver obverse.
When it was instituted, the CD was unique because it was awarded to all ranks, from the enlisted soldier to flag officers. Under the British system there had been a tradition of differentiating between those with long service in the ranks and those who were long-serving officers. Reserve officers in the army were typically awarded the Efficiency Decoration, known as the Territorial Decoration in the United Kingdom (officers in the regular army received no long service awards), while non-commissioned officers were awarded the Efficiency Medal (for reservists) or the Canadian Army Long Service and Good Conduct Medal (for members of the regular force). This situation was similar for the Royal Canadian Navy and the Royal Canadian Air Force, both of which possessed their own long service awards. Thus, under the previous system, reserve officers received long service "decorations" — and were therefore entitled to postnominals — while regular and reserve non-commissioned officers received long service "medals," which did not entitle the recipient to postnominals.
Canadian Forces' Decoration George VI issue obverse.
Canadian Forces' Decoration George VI issue reverse.
Another feature of the CD was that, with it, the period of service required to receive a long service award was standardized. Hitherto, the period of service varied depending on what service the recipient had spent the majority of his or her time in. A leading seaman in the RCN had to serve fifteen years to receive the RCN Long Service and Good Conduct Medal, while a leading seaman in the RCNVR only had to serve twelve years to receive the RCNVR Long Service and Good Conduct Medal. A similar situation existed in the other services where the period of service to qualify for a long service medal varied widely between ten and twenty years. This system of multiple awards was complex and laborious to administer.
Proposals for a uniquely Canadian long service award emerged from the Awards Coordination Committee in November 1946, though the idea originated shortly after the RCNR and RCNVR were unified into the Royal Canadian Navy (Reserve) and considered creating a single long service award for both the regular and reserve services.
Initially, two new awards were to be established: a Canadian Long Service Medal, to be awarded to regular members of the Canadian Armed Forces, and a Canadian Forces Long Service Decoration, to be awarded to those in the reserve force. Both were to be awarded for fifteen years of long service and good conduct. No differentiation was to be made on the basis of rank — only whether one was a member of the regular or reserve force. The Cabinet Defence Committee approved the CD on October 15, 1947, and designs were subsequently drawn up. By December 1949, plans for the Canadian Long Service Medal and Canadian Forces Long Service Decoration were merged into one award: the Canadian Forces' Decoration.
While it was now one award, this new plan still set out that a differentiation was to be made between those in the regular forces and those in the reserve. For recipients in the Regular Armed Forces, the insignia were to be made of silver gilt (gold in colour), as were the clasps. The ribbons were to be red with three thin, evenly spaced white stripes. For those in the Reserve Armed Forces, the insignia and clasps were to be of silver, while the ribbons were to be red with a single thin white stripe.
The plans for this differentiation were dropped in June 1950, however, and the original plan for the Regular Armed Forces CD was extended to the Reserve Armed Forces. One of the reasons why the Reserve Armed Forces CD was shelved was the perceived similarity between the ribbon of the reserve CD and the ribbon of the Order of the British Empire (military division), which was bordered on each side by thin grey stripes with a single grey stripe in the centre.
The CD was designed by Warrant Officer Barry Reddy. Early versions did not bear the King's effigy on the obverse or reverse. When he was presented with the first drawing in April 1948, George VI noted that the new decoration did not include this feature. This disappointed him, since he had always taken such a keen interest not only in the Canadian Armed Forces but also in Canada itself. The King's private secretary noted these "observations" and conveyed them to the governor general. The omission of the effigy was not done out of disrespect (his Royal Cypher was included), but once it was clear that the King had taken such an interest in the new decoration, a new design was created. Reddy redesigned the medal to include the King's effigy on the obverse. An earlier version bore a design of maple leaves, beavers, dams, and the aurora borealis.
The new award was announced on July 15, 1950, and most newspapers carried sketches of it. On June 7, 1951, at Rideau Hall, the first Canadian Forces' Decoration was presented to Viscount Alexander of Tunis, the governor general, by Brooke Claxton, the minister of national defence. At the same time nine other CDs were given to Lord Alexander for presentation to King George VI, Queen Elizabeth, Queen Mary the Queen Mother, Princess Elizabeth, Princess Margaret, the Duke of Gloucester, the Princess Royal, Lady Patricia Ramsay, and Princess Alice. All of these members of the Royal Family were serving members as colonels-in-chief of Canadian regiments.
After the death of King George VI in February 1952, the design of the CD was altered to bear the effigy of Queen Elizabeth II on the obverse while the reverse remained the same, save the removal of King George's Royal Cypher. Further alterations were also made to the medal: the suspender was changed from a bar bearing CANADA to a scroll-shaped opening at the top of the medal with a maple leaf at its base. The new method of suspension was designed by Bruce Beatty. The material from which the medal is struck was also changed from silver gilt to tombac. These changes were made in an effort to reduce costs. Some recipients of the King George VI CD also claimed that the suspension bar tended to break off, though there is little evidence of the medal being fragile.
While the King George VI CD bears the Royal Cypher on the reverse, the Queen Elizabeth II issue does not. When the first Queen Elizabeth II issue was struck, the Royal Canadian Mint accidentally used George VI's Cypher on the reverse. The problem was detected and the medals were destroyed. Subsequently, the Queen's Cypher (EIIR) was — also inadvertently — left off the reverse.
Since the presentation of the first CD to Viscount Alexander in 1951, it has been a tradition that all Canadian governors general receive the medal in their capacity as commanders-in-chief of Canada. Vincent Massey was the second governor general to receive the CD. Initially, to lend prestige to the new decoration, all members of the Royal Family who were members of the Canadian Armed Forces were automatically awarded the CD, even if they had not completed twelve years of service. Since 1953 the award has not been automatic to members of the Royal Family who serve as members of the Canadian Armed Forces, but required the normal twelve years of honorary service.
The Queen's own CD is of the King George VI type and is engraved HRH THE PRINCESS ELIZABETH. Prince Philip's CD is of the Elizabeth II type and is engraved HRH THE DUKE OF EDINBURGH. It was not until June 1954 that the first Elizabeth II type of the CD was struck.
Although it was originally planned to present the Canadian Forces' Decoration with a certificate signed by the governor general, this plan was never carried out and no official certificate is issued when the award is presented.
The Canadian Forces' Decoration
Canadian Forces' Decoration Elizabeth II issue obverse.
Canadian Forces' Decoration Elizabeth II issue reverse.
Origins: The Canadian Forces' Decoration was created as a Canadian replacement for a variety of British long service awards that were bestowed upon Canadians for service in the Royal Canadian Navy, the Canadian Army, and the Royal Canadian Air Force.
Criteria: Awarded to officers and non-commissioned members of the Canadian Armed Forces who have completed twelve years of service. The last eight years of claimed service must have been "good" in order to qualify for the CD.
Insignia: There have been two issues:
* • George VI type: Decagonal in shape, 36.5 mm in diameter, made from .800 silver, and gilded. The obverse bears an uncrowned effigy of King George VI and GEORGIVS VI D : G : BRITT : OMN : REX FID : DEF :. The reverse bears a naval crown at the top, three maple leaves, and an eagle, representing the three elements of the armed forces. The Royal Cypher is superimposed over this motif. The word SERVICE is embossed on a scroll at the base bordered on either side by a fleur-de-lys.
* • Elizabeth II type: Decagonal in shape, 36.5 mm in diameter, and made from tombac.[1] The obverse bears an uncrowned effigy of Queen Elizabeth II and ELIZABETH II DEI GRATIA REGINA with CANADA at the base. The obverse is identical to that of the King George VI issue, though without the Royal Cypher.
Suspender: Two types of suspenders have been used:
* • George VI type: A rectangular bar bearing CANADA attached to the medal by two scrolled arms connected to the actual medal by a single rivet.
* • Elizabeth II type: Two scroll bars with a maple leaf at the base through which the ribbon passes. The medal is struck from one piece of metal, whereas the George VI type was hand-assembled from two pieces.
Ribbon: Red, 38 mm wide, with three equally spaced white stripes 1 mm wide. Until 1967 the ribbon was a pale shade of red. Since then the red has been a deeper blood-red. Recipients of clasps wear a silver rosette on the undress ribbon for each clasp awarded.
Clasps: Awarded for an additional ten years of service, each bar measures 38 mm by 6 mm and there are three types:
* • Type I clasp: Plain .800 silver gilt and bears the 1921-style scalloped shield of the Royal Arms of Canada surmounted by a Tudor Crown.
* • Type II clasp: Plain gold in colour bearing the shield of the Royal Arms of Canada surmounted by a St. Edward's Crown. Two holes are drilled into each side of the bar so that it may be sewn onto the ribbon.
* • Type III clasp: Identical in appearance to the Type II clasp, with no holes drilled into the face of the bar, with a slip-on tang attached to the reverse, formerly used for investiture/presentation ceremonies and phased out in the 1990s.
Postnominals: CD.
Naming: Engraved capitals, abbreviated rank, initials, and surname of the recipient. For the George VI type these details are engraved on the reverse of the suspender, while the Elizabeth II issues are engraved on the rim in a variety of styles.
Canadian Forces' Decoration Type I clasp.
Canadian Forces' Decoration Type II clasp.
Canadian Forces' Decoration clasp for investitures.
Other: The Queen Elizabeth issues of the CD have been struck from four different types of metals. From 1955 to 1993 the CD was struck from tombac, an alloy composed of 12 percent zinc and 88 percent copper. This was a metal only readily available to the Royal Canadian Mint, and in June 1993 the metal was changed to gilding metal, which was composed of 95 percent copper and 5 percent zinc.[2] From 1991 to 2007 a number of other types of metals were used by some private manufacturers without the permission of the Department of National Defence. Certain manufacturers simply used whatever inexpensive metal could be obtained and hold a finish. Since the Royal Canadian Mint resumed striking the CD in 2007, the medal has been struck from gilding bronze.
The CD is awarded in a brown rectangular metal box whose lid is embossed with the Royal Arms of Canada. The interior is lined with black velvet and white or salmon-coloured satin. The original box was manufactured by the Farrington Manufacturing Company and was known as the Farrington Box.
Number: George VI type, approximately 14,000; Elizabeth II type, approximately 300,000; bars, approximately 104,500.
# 31
Exemplary Service Medals
Like the RCMP Long Service and Good Conduct Medal and the Canadian Forces' Decoration, the Exemplary Service Medals find their origins in the myriad of long service awards once used throughout the British Empire and Commonwealth. The closest relative of the Exemplary Service Medals is the RCMP Long Service Medal, which was established in 1934.
For nearly fifty years the RCMP Long Service Medal was the only standardized official police award and even then only members of the RCMP were eligible for it. Individual police forces developed a patchwork of awards, varying in criteria and design, none of which was officially recognized.
In 1909, Canadians were eligible for the King's Police Medal (see chapter 5), which was awarded for both gallantry and distinguished service. While the gallantry issue of the King's Police Medal was similar to the modern-day Medal of Bravery and Star of Courage, the distinguished service issue was more like the Order of Merit of the Police Forces or Meritorious Service Decorations. Although these two awards were at the disposal of the Canadian government, only fifty were ever awarded.
Other parts of the Commonwealth had established their own long service awards. The first was New Zealand, which instituted the New Zealand Police Long and Efficient Service Medal in 1868, more than twenty-five years before New Zealand became a country. This was the first official police long service medal to be issued with regularity. South Africa followed suit, establishing a prison long service medal and police long service medal in 1921 and 1922 respectively. Malta, Ceylon (Sri Lanka), and Cyprus also had special police long service medals by the late 1920s. One must therefore ask why it took so long for the various front-line services in Canada to call for the creation of their own awards. Newfoundland, then a separate dominion and not a province of Canada, established its own Constabulary Good Service Medal in 1892. British Columbia followed suit in 1946 with the establishment of the British Columbia Provincial Police (BCPP) Long Service Medal, which ceased to be awarded in 1950 when the BCPP was absorbed into the RCMP.
Newfoundland Constabulary Good Service Medal obverse.
British Columbia Provincial Police Long Service Medal obverse.
The answer can be found in the Canadian government's reluctance to award medals for civilian service, let alone to institute them, prior to 1967. Indeed, the dearth of Canadian long service medals was not for lack of trying. In 1958, for instance, the Canadian Association of Chiefs of Police asked the federal government if it would consider creating a Canadian Police Long Service Medal, and the reply was negative. The association then established its own award in 1960 — the Canadian Association of Chiefs of Police Long Service Medal — which was awarded for twenty years of good service, but the award was not recognized by the government and was considered only a temporary measure.
Throughout the 1970s, and especially following the establishment of the Canadian Bravery Awards in 1972, the Canadian Association of Chiefs of Police regularly asked the government to reconsider the creation of a Canadian Police Long Service Medal. In 1976, shortly before the tenth anniversary of the creation of the Order of Canada, the Cabinet was requested to consider the creation of such a medal. The proposal was finally accepted in 1980, and the Canadian Association of Chiefs of Police was invited to co-operate in the creation of the new award. Finally, on August 12, 1983, the Queen signed the letters patent creating the Police Exemplary Service Medal.
Although not initially part of the plan, the Exemplary Service Medals program eventually expanded to include correctional services, fire services, the Canadian Coast Guard, emergency medical services, and most recently peace officers. Prior to this, only the various fire services had been recognized and even then with unofficial awards that were provincially or locally sanctioned. The adoption of this series of exemplary service awards was rooted in part in a fear that each province and organization would institute its own awards with varying criteria and designs and then demand they be incorporated into the broader Canadian honours system. To avoid this, officials at the Chancellery and in the Privy Council Office encouraged the creation of a series of standardized national awards.
The Exemplary Service Medals are awarded only to those public services in which the public servant — whether it be police officer, corrections guard, ambulance driver, or firefighter — is working in potentially life-threatening environments. All of the medals require a minimum of twenty years of service in their respective organization. For each medal there is an advisory committee that submits lists of nominees to the governor general. These committees vary in composition. Whereas the advisory committee for the Police Exemplary Service Medal is the Decorations Committee of the Canadian Association of Chiefs of Police and is therefore national in scope, there are separate committees in each province and territory for the Corrections Exemplary Service Medal and Fire Service Exemplary Service Medal. The committees for the Canadian Coast Guard and emergency medical service versions of the medal, meanwhile, are more centralized, along the lines of the Police Exemplary Services Medal. These committees ensure that the minimum required service has been attained, and they advise the governor general on matters concerning the award of the medals.
Canadian Association of Chiefs of Police Long Service Medal.
The term Exemplary Service rather than Long Service was used in the criteria because the award is not granted automatically upon completion of twenty years of service: one must also have a good record. Exemplary service is defined as "service characterized by good conduct, industry, and efficiency that serves as a model for others." In extraordinary cases the medal may be awarded posthumously to persons who died in the performance of their duties. Until the late 1990s the names of those awarded the medal were published in the Canada Gazette. Each medal is awarded with a certificate signed by the governor general and bearing the recipient's name. Each bar awarded is also accompanied by a similar certificate.
All of the Exemplary Service Medals are of a similar design, featuring the maple leaf on the obverse and the Royal Cypher and Royal Crown on the reverse. The Canadian Coast Guard Exemplary Service Medal is slightly different in that it features a smaller maple leaf that is not cut out. With the exception of the Emergency Medical Services and Peace Officer Exemplary Service Medals, all of the Exemplary Service Medals were designed by Bruce Beatty.
These medals have all been made by a variety of manufacturers, including Rideau Ltée, Bond Boyd, Carat Médailles, and since 2013 the Royal Canadian Mint. Medals struck by the mint can be detected by the suspender, which is separate, and the metal is cupronickel, not rhodium-plated bronze or base metal as all the previous issues were.
Medals in this family have been presented in a variety of cases, generally a black or blue rectangular case embossed with the Royal Arms of Canada or the vice-regal lion. The interior of the case is lined with maroon or blue velvet and white satin. On very early issues the name of the manufacturer is stamped on the white satin. There are no postnominals associated with any of the Exemplary Service Medals.
The Police Exemplary Service Medal
Police Exemplary Service Medal obverse.
Origins: Largely based on the Royal Canadian Mounted Police Long Service Medal and by extension the Canadian Association of Chiefs of Police Long Service Medal.
Criteria: Awarded to serving police officers who have completed a minimum of twenty years of full-time service with one or more recognized Canadian police forces, including full-time police-cadet training, of such good standard as to warrant an award if such period of service, in whole or in part, has not been recognized by any other official, national long service and good conduct or efficiency decoration or medal. The award can be made retroactively if the person was a serving police officer on August 1, 1980, and meets the above-mentioned criteria. Members of the Royal Canadian Mounted Police who have been awarded the RCMP Long Service Medal and members of the Canadian Armed Forces Military Police who have received the Canadian Forces' Decoration are ineligible for the award. Bars may be awarded for ten years of additional service.
Insignia: A circular silver-coloured medal 36 mm in diameter with the obverse bearing a stylized maple leaf and the Scales of Justice circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown. Parts of the maple leaf not touching the inner edge of the medal are cut out. The very first lot of Police Exemplary Service Medals produced by the Royal Canadian Mint in 2013 was thicker than the earlier issues and lacked spines on the maple leaf displayed on the obverse. This design flaw was rectified after fewer than two hundred were struck.
Police Exemplary Service Medal reverse (standard for all Exemplary Service Medals except for the Canadian Coast Guard issue).
Royal Canadian Mint strike of the Police Exemplary Service Medal obverse.
Suspender: A footed fleur-de-lys attached to a straight bar.
Police Exemplary Service Medal certificate, first issue.
Ribbon: Five equal stripes — three blue and two yellow — the entire ribbon is 32 mm wide. The same as used for the Canadian Association of Chiefs of Police Long Service Medal.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Police Exemplary Service Medal certificate, current issue.
Naming: Engraved capitals with the initials and surname.
Number: 44,663.
The Corrections Exemplary Service Medal
Preliminary design for the Corrections Service Exemplary Service Medal.
Corrections Service Exemplary Service Medal obverse.
Origins: In part based on the Police Exemplary Service Medal. Earlier long service awards for corrections officials originated in South Africa with the creation of the Prisons Good Service Medal in 1921.
Criteria: Awarded to employees of the Canadian Correctional Service who have completed a minimum of twenty years of full-time paid service, not necessarily continuous, with one or more correctional services in Canada of such good standing as to warrant an award. Ten of those years must be spent as a peace officer in an institution, parole office, or probation office. No period of service that has been recognized by another long service medal shall be counted toward the award of the Corrections Exemplary Service Medal. Bars may be awarded for ten years of additional service. The recipient must have been an employee of the Canadian Correctional Service on or after June 11, 1984.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse bears a stylized maple leaf with a key and torch crossed over each other circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown. Parts of the maple leaf not touching the inner edge of the medal are cut out.
Suspender: A footed fleur-de-lys attached to a straight bar.
Ribbon: Composed of five equal stripes — three green and two yellow — the entire ribbon is 32 mm wide.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Naming: Engraved capitals with the initials and surname.
Number: 13,981.
The Fire Services Exemplary Service Medal
Fire Services Exemplary Service Medal obverse.
Origins: In part based on the Police Exemplary Service Medal. Prior to the establishment of the Fire Services Exemplary Service Medal, several provincial governments awarded medals to firemen for long service — most notably Ontario, Newfoundland, Nova Scotia, and British Columbia. The Fire Service Exemplary Service Medal is also similar in criteria to the Colonial Fire Brigade Long Service Medal created in 1934 and the British Fire Brigade Long Service and Good Conduct Medal established in 1954, as these were the first standardized long service awards for members of the fire services.
Criteria: Awarded to persons who have completed twenty years as a member of a fire service and whose service is of such good standard as to warrant the award. No period of service that has been recognized by another long service medal shall be counted toward the award of the Fire Service Exemplary Service Medal. Bars may be awarded for ten years of additional service. The recipient must have been a member of a fire service on or after August 29, 1985.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse shows a stylized maple leaf bearing a fire hydrant and crossed axes with a Maltese cross entered thereon, all of which is circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown. Parts of the maple leaf not touching the inner edge of the medal are cut out.
Suspender: A footed fleur-de-lys attached to a straight bar.
Ribbon: Composed of five equal stripes — three red and two yellow — the entire ribbon is 32 mm wide.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Naming: Engraved capitals with the initials and surname.
Number: 58,393.
The Canadian Coast Guard Exemplary Service Medal
Canadian Coast Guard Exemplary Service Medal obverse.
Canadian Coast Guard Exemplary Service Medal reverse.
Origins: In part based on the Police Exemplary Service Medal.
Criteria: Awarded to persons who have completed twenty years of service with the Department of Transport, ten years of which have been served with the Canadian Coast Guard in the performance of duties involving potential risk (as determined by the advisory committee). No period of service that has been recognized by another long service medal shall be counted toward the award of the Canadian Coast Guard Exemplary Service Medal. Bars may be awarded for ten years of additional service. The recipient must have been an employee of the Department of Transport on or after October 25, 1990.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse bears the crest of the Canadian Coast Guard (a single maple leaf in dexter and two dolphins sinister surrounded by a nautical rope tied at the base) circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown.
Suspender: A footed fleur-de-lys attached to a straight bar.
Ribbon: A 32 mm wide ribbon composed of seven stripes — a central 3 mm white stripe flanked on either side by a 3 mm blue stripe, a 7 mm yellow stripe, and another 3 mm blue stripe.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Naming: Engraved capitals with the initials and surname.
Number: 456.
The Emergency Medical Services Exemplary Service Medal
Emergency Medical Services Exemplary Service Medal obverse.
Emergency Medical Services Exemplary Service Medal certificate.
Origins: In part based on the Police Exemplary Service Medal. This is one of the first medals in the world to recognize long service rendered by emergency medical personnel.
Criteria: Awarded to persons who have completed twenty years of service with a recognized pre-hospital emergency medical service, ten years of which have been service in the performance of duties involving potential risk. The recipient must have been employed with an emergency medical service on or after October 31, 1991. No period of service that has been recognized by another long service medal shall be counted toward the award of the Emergency Medical Forces Exemplary Service Medal. Bars may be awarded for ten years of additional service.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse bears a stylized maple leaf superimposed with a star of life, which is in turn overlaid with a caduceus, all of which is circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown. Parts of the maple leaf not touching the inner edge of the medal are cut out.
Suspender: A footed fleur-de-lys attached to a straight bar.
Ribbon: A 32 mm wide ribbon with five equal stripes — three blue and two yellow. Superimposed over the centre of each blue stripe is a 2 mm orange stripe.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Naming: Engraved capitals with the initials and surname.
Number: 4,060.
The Peace Officer Exemplary Service Medal
Peace Officer Exemplary Service Medal obverse.
Origins: In part based on the Police Exemplary Service Medal.
Criteria: Awarded to persons who have completed twenty years of service with a recognized organization: Canadian Border Service Agency, Department of Citizenship and Immigration, Department of the Environment, Department of Fisheries and Oceans, Parks Canada Agency, and other federal or provincial organizations that employ peace officers. No period of service that has been recognized by another long service medal shall be counted toward the award of the Peace Officer Exemplary Service Medal. The recipient must have been employed as a peace officer on or after September 22, 1998. Bars may be awarded for ten years of additional service.
Insignia: A circular silver-coloured medal 36 mm in diameter. The obverse bears a shield set on a vallary star superimposed on a maple leaf, all of which is circumscribed with EXEMPLARY SERVICE — SERVICES DISTINGUÉS. The reverse bears the Royal Cypher surmounted by a Royal Crown. Parts of the maple leaf not touching the inner edge of the medal are cut out.
Suspender: A footed fleur-de-lys attached to a straight bar.
Ribbon: A 32 mm wide ribbon with a central green stripe 6 mm wide flanked on each side by 7 mm of yellow, 3 mm of blue, and 3 mm of dark blue.
Bars: A plain, straight silver bar bearing a maple leaf in the centre. A small silver maple leaf is worn on the undress ribbon to represent each bar awarded.
Naming: Engraved capitals with the initials and surname.
Number: 4,941.
# 32
Organizational Long Service MEDALS
The Canadian honours system includes two awards for service in non-governmental organizations that have a special relationship with the Crown. The Order of St. John has been active in Canada since 1883, has been recognized by a Royal Charter, and is included in the Canadian honours system (see chapter 16).
The Canadian Corps of Commissionaires has been active in Canada since 1925. Initially composed primarily of veterans of the First World War, it has grown to become one of the main security service providers in Canada, and many of its members continue to be former members of the Canadian Armed Forces. This close, though unofficial, relationship between the corps and the Canadian Armed Forces led to the creation of a long service medal for members of the corps in 1948. The corps has provided, and continues to provide, services to many Canadian Armed Forces bases and stations as well as government buildings, hospitals, and universities across Canada.
The Service Medal of the Order of St. John
Origins: The medal was originally approved in 1896 by Chapter-General, though it was not instituted until 1898. The final design was approved by Chapter-General on October 31, 1899. Originally, the medal could also be awarded for meritorious service, though this provision is no longer in effect. The Service Medal of the Order of St. John has been awarded to Canadians since its inception.
Criteria: Awarded to members of the St. John Ambulance organization for twelve years of service with a minimum of sixty hours of volunteer service per year. In other parts of the Commonwealth the medal has been awarded for ten or fifteen years of service.1
Insignia: A silver-coloured circular medal 38 mm in diameter. The obverse depicts a veiled effigy of Queen Victoria surrounded by the inscription VICTORIA D G BRITT REG F D IND IMP. The effigy of Victoria was taken from a bust sculpted by her daughter, Her Royal Highness Princess Louise, and this is the only medal in the Canadian honours system that bears the effigy of a deceased Sovereign and not the reigning monarch. The reverse of the medal depicts the shield of the Royal Arms of the United Kingdom within a garter bearing the motto HONI SOIT QUI MAL Y PENSE. Above this is a Tudor Crown and below it is the Prince of Wales's feathers. On either side is a badge of the order, each bearing a shield — the one on the left charged with the arms of the order, while the one on the right is charged with the arms of the Prince of Wales. The whole device rests upon a flourish of St. John's wort and is surrounded by the Latin words MAGNUS PRIORATUS HOSPITALIS SANCTI JOHANNIS JERUSALEM IN ANGLIA. Without question this is the most symbolically rich medal in the Canadian honours system.
Service Medal of the Order of St. John Type I with ring suspension obverse.
Suspender: Three types of suspenders have been employed:
* • Type I: A claw-footed suspender with a ring 15 mm in diameter.
* • Type II: A swivelling claw-footed suspender with a straight bar.
* • Type III: A non-swivelling footed suspender with a straight bar.
Service Medal of the Order of St. John Type III with bar suspension obverse.
Type I was used prior to 1913, Type II prior to 1965, and Type III since 1965.
Ribbon: A 38 mm wide ribbon watered and consisting of five equal stripes: three black and two white.
Service Medal of the Order of St. John certificate, obsolete.
Bars: There are two classes of bars:
* • A thin silver bar with sprigs of St. John's wort and a Maltese cross in the centre is awarded for each additional five years of service. A maximum of three five-year bars can be awarded.
* • A thin gold bar identical to the silver bar is awarded after twenty years of additional service. Each additional five years of service is recognized with a gold bar.
Service Medal of the Order of St. John certificate, current.
Postnominals: None.
Naming: The medals are currently issued unnamed. Prior to 1970, Canadian medals were named in hand-engraved capitals, number, abbreviated rank, initials, surname, and brigade.
Other: From 1898 through 1947 the medal was struck in silver; from 1948 until approximately 1970, in cupronickel; and since then in rhodium-plated base metal.
Number: 8,500 plus.
Service Medal of the Order of St. John reverse.
Gold oak leaf bar to the Service Medal of the Order of St. John.
Gold bar to the Service Medal of the Order of St. John.
Silver bar to the Service Medal of the Order of St. John.
The Canadian Corps of Commissionaires Long Service Medal
Obsolete bronze Canadian Corps of Commissionaires Long Service Medal.
Origins: The Corps of Commissionaires was created in 1859 in England as an organization dedicated to finding gainful employment for veterans of the Crimean War. Similarly, the Canadian Corps of Commissionaires was formed in Canada in 1925 with the purpose of finding employment for veterans of the First World War and later the Second World War. The Commissionaires continue to employ more than fifteen thousand Canadians.
A committee consisting of Major William Borrett, Colonel George Ross, and Brigadier-General Georges Francoeur designed the medal, and it was approved by the corps on June 4, 1947. At the request of the Canadian Corps of Commissionaires the secretary of state approved the creation of two long service medals on August 20, 1948. The bronze Corps of Commissioners Long Service Medal was awarded for ten years of service, while the silver medal was awarded for fifteen years. Recipients of the silver medal had to turn in their bronze ones.
The regulations were changed in 1998, and the Corps of Commissionaires Medal was brought into conformity with other Canadian long service awards. The bronze Long Service Medal was cancelled and the length of service for the silver medal was reduced to twelve years (the same period of service required to be eligible for the Canadian Forces' Decoration). The new Corps of Commissionaires Long Service Medal came into effect on February 26, 1998.
Awards of the medal are made by the national board of the Canadian Corps of Commissionaires on the recommendation of each division of the corps. The governor general has the authority to make extraordinary awards of the medal if the twelve-year period of service has not been completed for "special service to the Corps or for a record of exemplary service of such high standard as to merit the award of the medal."
Canadian Corps of Commissionaires Long Service Medal reverse.
Canadian Corps of Commissionaires Long Service Medal obverse.
Criteria: Awarded for twelve years of continuous or non-continuous service in the Canadian Corps of Commissionaires.
Insignia: On the current version of the medal, the obverse is surrounded by a garter-style ribbon bearing the motto VIRTUTE ET INDUSTRIA ("With courage and hard work"). At the centre is a natural maple leaf defaced with an anchor crossed with a sword and an eagle. The eagle was added in 1998 after a design change by Lieutenant-General René Gutknecht. Around the edge of the reverse are the words THE CANADIAN CORPS OF COMMISSIONAIRES — LE CORPS CANADIEN DES COMMISSIONNAIRES.
Suspender: An eyelet is affixed to the top of the medal and through this passes a small ring and a larger ring attached to the base of a top mounting bar that bears the words LABOR OMNIA VINCIT ("Work conquers all").
Ribbon: A 32 mm wide ribbon consisting of a central stripe of maroon bordered on either side by 6 mm of white bordered in turn by 3 mm of dark blue.
Bars: A silver bar adorned with a maple leaf is awarded for each five years of additional service. A maximum of three bars can be awarded to any one individual.
Postnominals: None.
Naming: Issued unnamed. Until the mid-1980s the medals were named on the reverse in a variety of styles.
Other: Until the early 1970s the silver Corps of Commissionaires Long Service Medal was struck in sterling silver by J.R. Gaunt of Montreal. Since that time the medals have been struck in base metal and have been rhodium-plated.
Number: Unknown.
# 33
Provincial Decorations and Medals
The inclusion of provincial honours in the national honours system is not a recent occurrence. However, there has been an increase in the number of officially recognized provincial decorations and medals in recent years. The most widely bestowed of these was the Centennial Medals issued by the Governments of Saskatchewan and Alberta. The status of the provincial orders as officially recognized honours that citizens are permitted to wear along with national honours has previously been examined (see chapter 17).
This chapter will only examine those provincial medals and decorations that have been officially recognized by the federal government. That is, they are listed in the national Order of Precedence for Canadian orders, decorations, and medals governed by a federal order-in-council. Provincial honours not included in the Order of Precedence are not recognized by the federal government and cannot be worn with official honours.
As with national honours, provincial honours can be divided into three categories: orders, decorations, and medals. Provincial honours include a number of bravery decorations, volunteer decorations, and commemorative medals. Unique among these are the Ontario Medal for Good Citizenship and the Saskatchewan Volunteer Medal, since these are specifically awarded for volunteerism and a similar honour does not exist at the national level. The closest national honour is the civil division of the Meritorious Service Decorations, and there is also the Sovereign's Medal for Volunteers.
There are also two obsolete provincial medals, namely, medals that have been replaced by federal honours. Both of these medals were specific to two provincial police forces — the old British Columbia Provincial Police, which was absorbed into the RCMP in 1950, and the Ontario Provincial Police, which continues to be responsible for many of the policing duties in that province.
Throughout the Commonwealth it was the practice for long-serving members of the constabulary to be rewarded with a medal for long service and good conduct. While Canada was quite late in establishing awards for police services, Newfoundland, then an independent dominion, was quick off the proverbial mark. The Newfoundland Constabulary Good Service Medal was established in 1892. The medal was awarded in silver and bronze and bore on the obverse a pair of crossed tipstaffs surmounted by a Royal Crown, the entire emblem circumscribed by the text NEWFOUNDLAND CONSTABULARY while the reverse of the oval medal simply bore the text FOR GOOD SERVICE. The medal could be awarded for long service or for a particularly meritorious act. The Newfoundland Constabulary Good Service Medal was never recognized as an official honour by the imperial government. The first Canadian medal of this type to be created was the Royal Canadian Mounted Police Long Service Medal, which was established by King George V in 1934.
Since there was no long service medal for members of police forces other than the RCMP, a number of provincial police forces established their own long service medals. In 1960 the Canadian Association of Chief of Police established its own long service medal, though this was never given official recognition by the Crown. The awarding of this medal ended in 1983 when the Police Exemplary Service Medal was created by the Queen.
A number of provinces also established long service medals for members of their municipal fire services. These, too, were never officially recognized by the Crown and were replaced by the Fire Service Exemplary Service Medal in 1985.
Certain provinces persist in the awarding of long service medals that duplicate a national award. For example, a firefighter would receive the Fire Exemplary Services Medal after twenty years and in at least one province if he or she served another five years — for a total of twenty-five years — the individual would be awarded a provincial long service medal for twenty-five years of service. This is known as dual recognition and runs contrary to one of the main principles of the Canadian honours system.
The Ontario Medal for Good Citizenship
Ontario Medal for Good Citizenship obverse.
Ontario Medal for Good Citizenship reverse.
Origins: On June 14, 1973, Premier William G. Davis of Ontario established the Ontario Medal for Good Citizenship, and three years later he also established two bravery awards, one for police officers and the other for firefighters. These were the first provincial honours created in Canada that would later be recognized by the federal government. The Ontario Medal for Good Citizenship was the first honour in Canada to specifically recognize voluntary service. Although some appointments to the Order of Canada have recognized voluntary service, this honour was devised solely to recognize that sort of service. Awards of this medal are made by the lieutenant governor on the advice of an independent advisory council. The medal was designed by the Toronto sculptor Gerald Gladstone. Nominations are solicited from the general public.
Criteria: Awarded to residents of Ontario who have through their exceptional long-term efforts made outstanding contributions to the well-being of their communities. Their services must have been given without expectation of remuneration or reward.
Insignia: A circular silver medal 36 mm in diameter whose obverse bears the stylized arms of the province and the text FOR GOOD CITIZENSHIP. On the reverse is a stylized Ontario trillium.
Suspender: A non-swivelling claw suspender with an ornate scroll bar.
Ribbon: A 38 mm wide white ribbon with a 3 mm wide gold stripe bordered on each side by 9 mm of white and 9 mm of green.
Bars: None.
Ontario Medal for Good Citizenship French obverse.
Naming: None.
Postnominals: OMC.
Other: The medal was originally struck in sterling silver by Birks, but this ceased in the mid-1980s and it is currently struck in silver-plated base metal.
Number: 325.
The Ontario Medal for Police Bravery
Ontario Medal for Police Bravery obverse.
Origins: Established in October 1975 by the provincial government through an order-in-council, this medal was created when the Canadian Decorations for Bravery were in their infancy and there was a perceived need for provincial bravery awards. The highly unusual design of the medal was created by Dora de Pedrey-Hunt.
Criteria: To recognize superlative acts of courage and bravery performed in the line of duty by members of Ontario's police services. Nominations are solicited from the general public.
Insignia: A gold-plated sterling silver cross with blue and white enamel bearing a stylized leaf pattern between each arm and the provincial logo in the centre. The uppermost arm of the cross is surmounted by a Royal Crown.
Suspender: The top of the Royal Crown is fitted with an eyelet through which passes a ring that is attached to another eyelet that is attached to a straight bar.
Ribbon: A 38 mm wide blue ribbon with two 3 mm wide gold stripes 5 mm from each edge of the ribbon.
Bars: None.
Naming: Engraved capitals with the initials and surname. Early awards also included the date of the incident being recognized.
Postnominals: None.
Number: 247.[1]
The Ontario Medal for Firefighter Bravery
Ontario Medal for Firefighter Bravery obverse.
Origins: Established September 1976 after the creation of the Ontario Medal for Police Bravery, this medal was created when the Canadian Decorations for Bravery were in their infancy and there was a perceived need for provincial bravery awards. As with the Ontario Medal for Police Bravery, the design was created by Dora de Pedrey-Hunt. Nominations are solicited from the general public.
Criteria: To recognize superlative acts of courage and bravery performed in the line of duty by members of Ontario's fire services.
Insignia: A gold-plated sterling silver cross with red and white enamel bearing a stylized flame pattern on each arm and the provincial logo in the centre. The uppermost arm of the cross is surmounted by a Royal Crown.
Suspender: The top of the Crown is fitted with an eyelet through which passes a ring that is attached to another eyelet that is attached to a straight bar.
Ribbon: A 38 mm wide red ribbon with 2 3 mm wide gold stripes 5 mm from each edge of the ribbon.
Bars: None.
Naming: Engraved capitals with the initials and surname. Early awards also included the date of the incident being recognized.
Postnominals: None.
Number: 201.[2]
The Saskatchewan Volunteer Medal
Saskatchewan Volunteer Medal obverse.
Origins: This medal was largely based on the Ontario Medal for Good Citizenship and is the second Canadian medal created specifically for volunteer activity. The concept for the medal was devised by D. Michael Jackson, the province's chief of protocol. It was created in 1995 on the occasion of the province's ninetieth anniversary of entering Confederation. The medal was created under the Provincial Honours and Emblems Act. During the 2005 Royal Visit to Canada, Her Majesty the Queen signed the official design of the medal, making the inclusion of the Crown on the insignia official.
It is considered to be junior to the province's order — the Saskatchewan Order of Merit. The medal was designed by Jack Glasser and is struck by Rideau Ltée. Awards are made by the lieutenant governor on the advice of the Saskatchewan Honours Advisory Council.
Criteria: Awarded to residents and former residents of the province who have provided, without reward or gain, outstanding volunteer service or exceptional community involvement in an area beyond the performance of the individuals' normal duties or the exercise of the responsibilities of a profession to which that individual belongs.
Insignia: A silver-coloured medal 36 mm in diameter, the obverse depicting the provincial shield surmounted by a Royal Crown, all enamelled, and circumscribed by the text NOS IPSOS DEDIMUS SASKATCHEWAN ("We gave ourselves"). The reverse is plain.
Suspender: The letter V is attached to the top of the disc, which is attached to a straight bar.
Ribbon: A 32 mm ribbon divided into 3 equal portions of green, yellow, and green.
Bars: None.
Naming: Engraved with a three-digit issue number.
Postnominals: SVM.
Other: The medal is made by Rideau Ltée out of plated base metal.
Number: 143.
The Saskatchewan Centennial Medal (2005)
Saskatchewan Centennial Medal obverse.
Saskatchewan Centennial Medal reverse.
Origins: To recognize the centennial of the Province of Saskatchewan entering Confederation, the provincial government decided to strike a commemorative medal to mark the occasion. The concept for the medal was developed by D. Michael Jackson, the province's then long-serving chief of protocol. The design and criteria were loosely based on other similar commemorative medals, notably the 125th Anniversary of the Confederation of Canada Medal.
Criteria: Criteria for this medal are a broad range of contributions to the community and to the province through leadership, voluntarism, community involvement, and outstanding personal achievements. Recipients were nominated by federal MPs from Saskatchewan, members of the Provincial Legislative Assembly, a number of volunteer organizations, and the lieutenant governor.
Saskatchewan Centennial Medal certificate.
Insignia: A circular bronze medal 36 mm in diameter. The obverse depicts a Royal Crown in the centre atop three prairie lilies with the provincial motto on a scroll below, the obverse circumscribed by the text SASKATCHEWAN 1905–2005. The reverse bears the provincial shield on a plain matte finished field. The medal was designed by Warwick Communications of Saskatoon. Non-residents of Saskatchewan were also eligible for the medal, though fewer than fifty were awarded.
Suspender: A footed suspender with a straight bar.
Ribbon: Dark gold, 32 mm wide, with a 2 mm green stripe inset 2 mm from each edge of the ribbon.
Bars: None.
Naming: None.
Postnominals: None.
Other: The Saskatchewan Centennial Medal was awarded with a certificate signed by the lieutenant governor and the premier. The medal was struck by Bond Boyd Ltd.
Number: 4,000.
The Alberta Centennial Medal (2005)
Alberta Centennial Medal obverse.
Alberta Centennial Medal certificate.
Origins: The medal was devised as a mechanism to recognize outstanding citizenship in the province and to commemorate the entrance of Alberta into Confederation. The concept was largely based on the Saskatchewan Centennial Medal, which is largely based on the 125th Anniversary of the Confederation of Canada Medal awarded in 1992.
Criteria: Awarded to people who made a significant contribution to their fellow citizens, their community, and to Alberta. Recipients were nominated by federal MPs from Alberta, members of the Provincial Legislative Assembly, two hundred volunteer organizations, and the lieutenant governor. When the medal was established, it was only open to residents of the province, though this was changed to allow non-residents to receive the honour. Fewer than one hundred were presented to non-residents.
Insignia: A gold-plated bronze medal 36 mm in diameter. The obverse bears the full provincial coat of arms and is circumscribed by the text ALBERTA CENTENNIAL 1905–2005. The reverse bears a large Canadian maple leaf with the provincial shield in the centre circumscribed with the text HONOURING OUTSTANDING ALBERTANS. The medal was designed by John Smith of Artsmith Communications.
Suspender: A straight graduated silver suspender 38 mm in width at its widest point is attached to the top of the disc via a link and eyelet that is soldered on the top of the disc.
Ribbon: A ribbon 32 mm wide edged on each side with 3 mm of blue, 3 mm of gold, 2 mm of green, and 2 mm of white, with a central 1 mm stripe of pink. The blue represents clear skies and sparking lakes; gold, wheat fields and resource wealth; green, forests; white, mountains; and pink, the wild rose, which is the provincial flower.
Bars: None.
Naming: None.
Postnominals: None.
Other: The medal was presented with a certificate signed by the lieutenant governor and the premier.
Number: 8,000.
The British Columbia Firefighter Medal for Bravery
British Columbia Firefighter Medal for Bravery obverse.
Origins: This medal was modelled on the Ontario Medal for Firefighter Bravery. The medal was established on June 8, 1988, by the provincial fire commissioner. Nominations are made to the provincial fire commissioner whose office researches each nomination. Recommendations for award of the medal are made by the Bravery Award Committee to the minister of public safety and the solicitor general. The medal was formally included in the province's Fire Services Act in 1999 along with the British Columbia Fire Service Long Service Medal.
Criteria: Awarded to B.C. firefighters who have performed acts of courage and bravery without concern for their personal safety.
Insignia: A silver coloured, 38 mm wide, maple leaf–shaped medal enamelled red with smaller silver maple leafs between the points, the central red maple leaf bearing in the centre a disc containing a firefighter's cross superimposed with the full provincial arms in gold contained in a square surrounded by a gold wreath of laurel leaves at the base of which appears the text MEDAL OF VALOUR on a scroll.
Suspender: A small ring is attached to the uppermost point of the maple leaf through which passes another ring that attaches to a straight suspender bar.
Ribbon: A 32 mm wide ribbon with a central 8 mm stripe of red bordered on each side by 8 mm of white and 4 mm of red.
Bars: None.
Naming: None.
Postnominals: None.
Number: Unknown.
The British Columbia Fire Service Long Service Medal
British Columbia Fire Service Long Service Medal obverse.
British Columbia Fire Service Long Service Medal reverse.
Origins: This medal was established on June 8, 1988, by the provincial fire commissioner. The medal was formally included in the province's Fire Services Act in 1999 along with the British Columbia Firefighter Medal for Bravery.
Criteria: Awarded to paid and volunteer firefighters who have completed twenty-five years of paid or volunteer fire service in British Columbia. Bars are awarded for ten years of additional service.
Insignia: A gold medal 36 mm in diameter, the obverse depicting a firefighter's cross with the full provincial arms in the centre, the entire emblem circumscribed by the text BRITISH COLUMBIA FIRE SERVICE ACT.
Suspender: A small eyelet is attached to the top of the disc through which passes a ring that is attached to a bar and the ribbon. The bar has decorative finials attached to each outer end.
Ribbon: A 32 mm wide ribbon with an 8 mm central stripe of yellow bordered on each side by 8 mm of red and 4 mm of blue.
Bars: A 32 mm wide gold rectangular bar bearing the full provincial arms in the centre is awarded for ten years of additional service.
Naming: None.
Postnominals: None.
Number: Unknown.
The Newfoundland and Labrador Award for Bravery
Newfoundland and Labrador Award for Bravery obverse.
Origins: In 1893, Sir Terence O'Brien, governor of Newfoundland, established a bravery award styled the Newfoundland Silver Star for Bravery at Sea. While the Crown never officially recognized it, the medal continued to be awarded into the period following the end of the Great War. In 1984 the provincial House of Assembly established a new honour — the Newfoundland and Labrador Award for Bravery — which was subsequently recognized by the national honours system. Recommendations for the award are made by a review panel made up of the chief justice of the province, the president of Memorial University, the chief superintendent of the RCMP in the province, the chief of the Royal Newfoundland Constabulary, the private secretary to the lieutenant governor, and the clerk of the Executive Council. The medal is struck by the Royal Canadian Mint and was designed by Ian Stewart, CM.
Criteria: The award is presented to citizens of the province whose lives and safety have been risked while giving assistance to other people.
Insignia: A gold-plated bronze medal 57.2 mm square, the obverse containing a circular inner rim depicting a stylized drama at sea "symbolic of the courageous heritage of Newfoundland and Labrador," while the reverse is plain bearing the text NEWFOUNDLAND AND LABRADOR FOR BRAVERY.
Suspender: A loop is attached to the top of the insignia through which the neck ribbon passes.
Ribbon: Light blue and 32 mm wide.
Bars: None.
Naming: Engraved capitals, initials, and surname.
Postnominals: None.
Other: The medal is worn around the neck similar to an Olympic medal.
Number: 45.
The Newfoundland and Labrador Volunteer Service Medal
Origins: In 1996 the government of Newfoundland and Labrador established the Newfoundland and Labrador Volunteer Service Medal, which was roughly based on the Saskatchewan Volunteer Medal. The medal was created by an act of the provincial legislature that received Royal Assent on December 19, 1996. To recognize the International Year of Volunteers, the medal was presented in 2001. Although the name of the medal was formally the Newfoundland and Labrador Volunteer Service Medal, this was shortened to simply the Newfoundland and Labrador Volunteer Medal to help avoid confusion with the Newfoundland and Labrador Volunteer War Service Medal, which had been awarded for military service rendered during the Second World War.
Recipients were selected by a selection committee made up of seven people appointed by the lieutenant governor-in-council. Nominations were received from across the province throughout 2001, and the first medals were presented by Lieutenant Governor A.M. House in Happy Valley–Goose Bay on November 19, 2001. Although the medal was intended to be awarded on an annual basis, it was only awarded in 2001 and no awards have been made since that time. In many ways this medal can be considered a "commemorative" medal, since it was awarded for one year only. The medal was designed by Ian Stewart, CM, and was struck by the Royal Mint in London.
Criteria: Awarded to long-time residents of Newfoundland and Labrador who provided volunteer services for a minimum of ten years to their community without reward or gain in an area beyond their normal duties or responsibilities. A total of ninety-six medals were awarded, two to citizens in each of the province's electoral districts — one woman and one man. Members of the provincial House of Assembly, House of Commons, Senate, and judges were not eligible to receive the medal.
Insignia: A silver-coloured circular medal 36 mm in diameter. The obverse depicts the shield of arms of the Province of Newfoundland and Labrador circumscribed by the text NEWFOUNDLAND AND LABRADOR VOLUNTEER SERVICE MEDAL. The reverse has the text NON SIBI SED ALIIS ("Not for oneself but for others").
Suspender: A claw-footed suspender with an ornate scroll bar.
Ribbon: A 32 mm wide maroon ribbon with a central 8 mm stripe of white bordered on each side by 2 mm of green.
Bars: None.
Naming: Unknown.
Postnominals: None.
Other: A complete roll of the recipients of this medal hangs in the main hall of the Confederation Building, the province's House of Assembly. In place of a certificate, recipients of the medal were presented with a wood-and-metal plaque.
Number: 96.
# 34
Commonwealth and Foreign Awards
The bestowal of Commonwealth and foreign awards upon Canadians has periodically been the source of public discussion. The best-known case in this respect was that of Lord Black of Crossharbour. The British government wished to have Conrad Black, then the owner of the Daily Telegraph, appointed as a life peer of the United Kingdom and thus a member of the British House of Lords. Although Black was at the time a dual citizen, the Canadian government objected to the proposal. After a lengthy court battle, Black decided to renounce his Canadian citizenship so that he could accept the peerage.
The practice of governments seeking permission from one another to bestow honours upon foreign nationals is neither new nor unique to Canada. Throughout the Commonwealth there have long been regulations requiring such permission. During the two world wars, the British government gave unrestricted permission for British subjects to receive foreign awards from allied nations. While one could receive a particular honour, special permission had to be received from the King to wear it. Unrestricted permission for wear was often given, with the foreign decorations being worn in order of the date of award and after all British medals.
During peacetime, the norm has been for more restricted permission to be granted; that is, the recipient may only wear the particular honour while in the country from which it was granted (or its embassy), or while the head of state of the particular country is on a visit to Britain. In Canada the same practice was followed during the First World War, but things changed during the Second World War. Foreign awards had to be approved by the Awards Coordination Committee, though there are few instances of the committee refusing to allow a particular award to be granted. Indeed, the honours were often conferred before the committee had even learned about them!
First and Second World War group to H.H. Leather, head of the Canadian Red Cross Society during the Second World War (MBE, Officer of the Order of St. John, 1914–18 British War Medal, Victory Medal, Queen Elizabeth II Coronation Medal, Officer of the Order of the Red Cross of Belgium, Bulgaria Red Cross Decoration, Greek Red Cross Medal of Recognition 1st Class, Officer of the Order of the Red Cross of the Netherlands).
Following the end of the Second World War, Canada reverted to a policy of not allowing its citizens to accept any foreign awards, though exceptions were made for Prime Minister William Lyon Mackenzie King, who was awarded four foreign honours: the Grand Cross of the Légion d'honneur from France, the Grand Cross of the Order of the Oak Crown from Luxembourg, the Grand Cross of the Order of Leopold I from Belgium, and the Grand Cross of the Order of the Netherlands Lion. It is ironic that while Mackenzie King imposed the most stringent restrictions upon other Canadians accepting British and foreign awards, he remains to this day the most honoured prime minister in Canadian history!
Second World War group to Captain G.G. Armour (1939–45 Star, France and Germany Star, Defence Medal, Canadian Voluntary Service Medal and clasp, 1939–45 War Medal with MID, Efficiency Medal, French Croix de Guerre).
By 1956 a new policy was instituted, one that allowed Canadians to accept foreign honours only if they were awarded in connection with saving or attempting to save a life. Canadians were officially prohibited from receiving foreign awards for other reasons, though the policy was only sporadically enforced.
A unique breach in this policy occurred in the spring of 1967 when a Second World War veteran of the Royal Canadian Air Force, Robert Norman Thompson, was made a Grand Cross of the Order of the Star of Ethiopia for his service to that country from 1944 to 1958 as founding commander of the Imperial Ethiopian Air Force and as deputy minister of education.
The Right Honourable John Diefenbaker wearing his Order of the Companions of Honour.
Thompson, at that time an MP who had recently served as leader of the Social Credit Party of Canada, attended a dinner at Rideau Hall honouring Emperor Haile Selassie during the emperor's visit to Canada for the Centennial. Selassie chided Thompson for not wearing his Ethiopian decorations to the event. Thompson replied that he possessed none, so the emperor decided to make the award on the spot.
Policy Respecting the Awarding of an Order, Decoration, or Medal by a Commonwealth or Foreign Government, January 1988
1. Any Commonwealth or foreign government desiring to award an order, decoration, or medal to a Canadian citizen shall obtain prior approval from the Government of Canada.
2. Commonwealth and foreign governments desiring to award orders, decorations, or medals to Canadian citizens shall submit their proposals to the Government of Canada through their diplomatic missions in Canada.
3. The Government of Canada will consider granting the approval referred to in section 1 for the awarding of an order, decoration, or medal offered in recognition of:
> (a) an extraordinary service to mankind;
>
> (b) conspicuous bravery in saving or attempting to save a life;
>
> (c) any exceptional service rendered to the country desiring to make the award; or
>
> (d) any substantial act or acts contributing to better relations between Canada and the country desiring to make the award.
>
> 4. The Government of Canada shall not grant approval referred to in Section 1 for an award:
>
> (a) that is at variance with Canadian policy or the public interest;
>
> (b) that carries with it an honorary title or confers any precedence of privileges;
>
> (c) that is conferred otherwise than by a head of state or a government recognized as such by Canada;
>
> (d) that is conferred in recognition of services by an individual in the employ of Her Majesty in Right of Canada or of a province in the normal performance of official duties; or
>
> (e) that is in respect of events occurring more than five years before the offer of the award.
>
> 5. The Honours Policy Committee may advise and recommend on the interpretation and application of this policy on the disposition of cases arising therefrom.
# PART III
# 35
The Memorial Cross and Next-of-Kin Mementos
Since the end of the First World War, Canada has instituted a number of mementos for the next of kin of those killed while on active service. The most prominent of these is the Memorial Cross, which has been joined by other items such as the Memorial Bar, the RCMP Memorial Memento, the Memorial Scroll, and most recently the Memorial Ribbon.
The Memorial Cross is in many respects the first uniquely Canadian award, that is, one that was instituted independently of any previously existing award and one that applied to Canadians alone. The creation of the Memorial Cross was formally announced by Major-General Sydney Chilton Mewburn, minister of militia and defence, in the House of Commons on March 23, 1919, and was later formalized by an order-in-council. Mewburn had initially created the Memorial Cross under the War Measures Act, but it was later thought more prudent to regulate the award through an order-in-council and not through the Routine Orders of the Canadian Expeditionary Force alone. The nascent idea for the Memorial Cross originated with William Alexander Fraser, a noted Canadian author and writer. Fraser wrote an article in the Toronto Star on September 23, 1916, entitled "Cross for Mothers," which suggested that an appropriate symbol be developed to recognize the mothers of Canada's fallen soldiers.
The Memorial Cross has remained largely unchanged since its institution on December 1, 1919. Along with the Order of St. John, RCMP Long Service Medal, Canadian Forces' Decoration, and a few other honours, it is one of the few distinctions carried over from the pre-1967 honours system.
During the First World War, the Memorial Cross was issued to the widows and mothers (if living) of every fallen soldier, sailor, or airman. In cases where the widow or mother had died prior to receiving the cross, it would go to the father or eldest next of kin. The minister of militia and defence could also designate another recipient. The cross was even awarded to Newfoundlanders a full generation before the dominion joined Confederation.
New Zealand Memorial Cross.
The primary criteria for the award was that the soldier died in or as a result of service rendered during the First World War while a member of one of His Majesty's armed services. Thus the mother of a Canadian who died while serving in the New Zealand Expeditionary Force would receive the cross. Similarly, if the family applied, provisions were made to allow Canadians who died in or as a result of service in other military forces to be issued the cross. Through this provision there are a few examples of Canadian mothers or wives receiving the Memorial Cross when their sons or husbands died while serving in the United States Army during the First World War.
The criteria for the Second World War Memorial Cross were almost identical to those used during the First World War, with the simple addition of eligibility of next of kin to airmen, merchant seamen, and members of the Corps of Canadian Fire Fighters for Service in the United Kingdom.
The symbol of the Memorial Cross is a familiar one to many Canadians, since a large replica of the Memorial Cross hangs in the Memorial Chamber of the Peace Tower in Parliament. Similarly, large versions of the King George V, King George VI, and Queen Elizabeth II Memorial Crosses were cast in bronze and used on three of the four corners of the tomb of Canada's Unknown Soldier. Every year a "silver cross mother" lays a wreath at the National War Memorial in Ottawa on behalf of all mothers and wives who have lost a loved one. The Memorial Cross insignia is based on a three-quarter version of the Military Cross, with the addition of maple leaves on three of the four arms of the cross and a wreath that passes between each of the cross's arms.
The Memorial Cross was viewed as such a useful memento of remembrance that the practice was adopted by New Zealand. On September 12, 1947, King George VI approved the establishment of the New Zealand Memorial Cross. The criteria and design for this award are almost identical to those for the Canadian Memorial Cross. In place of the maple leaves at each end, however, New Zealand used fern leaves. The ribbon was purple. Like Canada, New Zealand continues to award the Memorial Cross, expanding the criteria to include peacekeeping and other international missions.
Since 2003 the criteria for the Memorial Cross has been broadened, primarily as a result of Canada's involvement in Afghanistan. In May 2006, Captain Nichola Goddard was killed while serving with the Canadian Armed Forces in Afghanistan. She became the first female combat casualty in Canadian history, and her husband, Jason Beam, became the first husband presented with the Memorial Cross. Since 2007 members of the Canadian Armed Forces are requested to submit the names of three people who would be presented with the Memorial Cross in the event of death while serving in the Canadian Armed Forces. Members are not required to list spouses or parents and can choose friends to receive the Memorial Cross.
United Kingdom Elizabeth Cross.
In July 2009 the United Kingdom established the Elizabeth Cross, which is based on the Canadian Memorial Cross in terms of the criteria and actual design. In place of maple leaves on each of the arms, there are a rose for England, a thistle for Scotland, a shamrock for Ireland, and a daffodil for Wales. Another slight differences is the method of suspension, which for the Elizabeth Cross consists of a brooch on the reverse of the actual cross in place of a stylized mounting bar.
The Memorial Cross (George V)
Memorial Cross (George V) obverse.
Criteria: Issued as a memento of personal loss and sacrifice on the part of the mother and widow (or eldest next of kin) of a sailor or soldier of Canada or Newfoundland who laid down his life for his country during the First World War or who died from causes attributable to service during the First World War.
Insignia: A silver cross pattée equally proportioned and 32 mm in width. The cross is superimposed on a wreath of laurel. The uppermost arm is surmounted by a Tudor Crown, while the remaining three bear a maple leaf. The centre of the cross displays the Royal Cypher (GRI) in raised letters. The reverse of the cross is plain, while the lower arm of the cross is normally stamped with a maker's mark and STERLING.
Suspender: A small silver eyelet is mounted on the uppermost arm through which passes a loop and the ribbon.
Ribbon: An ecclesiastical purple ribbon 10 mm wide and of varying lengths. The King George V issue Memorial Cross was worn around the neck.
Naming: Engraved capitals, number, abbreviated rank, initials, and surname of the officer or soldier commemorated.
Other: The Memorial Cross was issued in a black case embossed with a Tudor Crown on the lid. The interior was of white satin and velvet. The manufacture and engraving of the Memorial Crosses were undertaken by a variety of firms.
Number: 65,000 plus. No precise figure exists, since the number grew with time as more and more veterans of the Great War died as a direct result of illness or wounds attributed to the war.
The Memorial Cross (George VI)
Memorial Cross (George VI) obverse.
Memorial Cross (George VI) reverse.
Criteria: Issued as a memento of personal loss and sacrifice on the part of the mother and widow (or in certain circumstances the father or eldest next of kin) of a sailor, soldier, airman, merchant seaman, or firefighter of Canada or Newfoundland who laid down his life for his country during the Second World War, or who dies or died from causes attributable to service rendered during the Second World War.
Insignia: A silver cross pattée equally proportioned and 32 mm in width. The cross is superimposed upon a wreath of laurel. The uppermost arm is surmounted by a Tudor Crown, while the remaining three bear a maple leaf proper. The centre of the cross displays the Royal Cypher (GVIR) in raised letters. The reverse is plain with the lower arm stamped with a maker's mark and STERLING.
Suspender: A small silver eyelet is mounted on the uppermost arm through which passes a loop and the ribbon. After January 1945, the Memorial Cross was mounted on a silver bar 32 mm long and 5 mm wide with flared edges, bearing maple leaves (in the same style as on the cross) at each edge.
Ribbon: An ecclesiastical purple ribbon 10mm wide and of varying lengths. The King George VI issue Memorial Cross was worn around the neck. No ribbon was issued after January 1945.
Naming: A variety of styles depending on the branch of service:
* • Navy: Engraved on two lines — initials, surname, abbreviated rank, and RCN, RCNR, or RCNVR.
* • Army: Engraved on three lines — number, abbreviated rank, initials, and surname.
* • RCAF: Engraved on three lines — abbreviated rank, initials, surname, and number.
Other: The Memorial Cross was issued in a black case embossed with a Tudor Crown on the lid. The interior was of white satin and velvet.
Number: 60,000 plus.
The Memorial Cross (Elizabeth II)
Memorial Cross (Elizabeth II) obverse.
Memorial Cross (Elizabeth II) reverse.
Criteria: Issued as mementos of personal loss and sacrifice on the part of mothers and widows of personnel of the Canadian Armed Forces who have laid down their lives for their country while in or proceeding or returning from an area of hostilities outside Canada or who die of causes attributable to service in such areas in consequence of any action undertaken by Canada under the United Nations Charter, the North Atlantic Treaty, or any other instrument for collective defence that may be entered into by Canada. Since January 1, 2007, all service-related deaths qualify for award of the Memorial Cross provided the person served in the Canadian Armed Forces after October 6, 2001.
Insignia: A silver cross pattée equally proportioned and 32 mm in width. The cross is superimposed upon a wreath of laurel. The uppermost arm is surmounted by a Royal Crown, while the remaining three bear a maple leaf proper. The centre of the cross displays the Royal Cypher (EIIR) in raised letters. The reverse of the cross is plain, the lower arm being stamped with a maker's mark and STERLING.
Suspender: A small eyelet is mounted on the uppermost arm through which passes a loop that is attached to another eyelet that is in turn mounted on a bar 32 mm long and 5 mm wide with flared edges, bearing maple leaves (in the same style as on the cross) at each edge.
Ribbon: None.
Naming: Engraved capitals, number, abbreviated rank, initials, and surname. Occasionally, the branch of service is also listed in an abbreviated form.
Other: The Memorial Cross was issued in a rectangular black leatherette case embossed on the outer lid with the Royal Arms of Canada surrounded by a border of gold. The interior is lined with black velvet and off-white satin. For a period in the 1990s up to 2002, the Memorial Cross was issued in a red burgundy case similar in design to the case Canadian Forces' Decorations were presented in at the time. As of 2003, the case of issue has been impressed with a simple Royal Crown in gold with no border. Since 2005 the Memorial Cross has also been issued with a small message printed on a square card noting the significance of the insignia.
Number: 2,000 plus. Approximately one thousand crosses were granted for the Korean War, and the balance for soldiers killed in United Nations Peacekeeping Missions — most recently for those killed in the Southwest Asia (Afghanistan) operations. While an official award, the Memorial Cross is not ranked in the Order of Precedence, and it is only since the Korean War that accurate records of the number of crosses issued have been maintained.
The Memorial Bar (Birks Memorial Bar)
First World War Memorial Plaque.
The Memorial Bar, often referred to as the "Birks Bar," can appropriately be compared to the Memorial Plaque presented to next of kin following the Great War. The Memorial Bars were presented to the next of kin of Canadian service personnel who died during the Second World War up to 1947.
Canada, like Britain, did not name its Second World War medals, mainly because of the cost, logistics, and time required to name what typically amounted to more than three medals for each veteran.[1] The medals and stars were struck at the Royal Canadian Mint in Ottawa using punches manufactured by the Royal Mint in London.[2] It was not until 1948 that the striking of Second World War medals was at full capacity, and in September 1949 the Royal Canadian Mint in co-operation with the Department of National Defence began mailing out tens of thousands of medals to those who had earned them.[3] The receipt of the medals years after the conflict was a source of frustration among many veterans' groups as was the fact that the medals were issued unnamed.
Memorial Bar mounted on a Second World War group.
Memorial Bar modern issue in French.
Memorial Bar modern issue in English.
The failure to name medals caused a debate in the Canadian House of Commons. The attack on the government's failure to name the medals was led by Conservative MP Major-General George Pearkes. Pearkes began by questioning Liberal MP and Minister of Veterans Affairs Brigadier Milton F. Gregg. The entire spectacle must have been something to watch from the gallery of the Commons — a pair of men, both holders of the Victoria Cross, arguing over medals in Parliament!
On April 21, 1950, following a number of media reports and statements by members of the Canadian Legion (later retitled the Royal Canadian Legion), Pearkes rose in the House of Commons during the budget debate and grilled his fellow VC holder on the government's failure to name Second World War medals:
> General dissatisfaction has been expressed because men who were entitled to receive service medals find their names have not been inscribed on the medals. The medals issued after World War I gave the man's name and his regimental number on the edge of the medal, but in this war the medals have been issued without any name whatever and they can be pawned off and passed around without a man being able to say, "This is the medal I earned."[4]
Gregg did not reply to Pearkes's line of questioning, but he did set his department to work on finding a solution. Pressure was mounting from veterans' groups, and before plans to remedy the situation could be set, the question was again raised in the Commons. This time Howard Green, another Conservative MP and veteran of the First World War, rose to address the issue: "I suppose it would be too late now to do anything about the medals for the men returned, but I believe the Minister should give consideration to calling in the medals which were sent out to the next of kin and having the names inscribed on them."[5]
The debate had moved from the massive logistical undertaking of having all Second World War service medals named to one of just naming those of casualties. Green went on to use the ever-effective parliamentary tactic of reading a letter from a concerned constituent:
> If these medals had borne my son's name, I would have been glad to receive them and proud to place them in the family archives. It happened that my son died well. But they do not bear his name. They are just four medals picked from four stock-piles, packed, not too tidily, into four little boxes and sent with instructions on how they should be worn. Worn by whom? I suggest the government could, even now, cleanse from the national record the more hurtful feature of its disgraceful parsimony, by recalling at least all of the medals sent to next of kin and having them inscribed with the names of the men who earned them.[6]
With pressure growing from the general public and veterans' groups, the government was unable to ignore the situation any longer. The recall of every medal issued for naming was impractical, and the Royal Canadian Mint advised that naming nearly four million medals would double the delivery time and cause a backlog, especially considering the return of medals already issued.
The Department of Veterans Affairs developed a proposal for "a silver bar of identification," an engraved silver bar that could be sewn onto the ribbons and made suitable for framing by the next of kin. By the end of July, the proposal was complete and it was presented to the Cabinet on August 18, 1950. At this point Gregg had been replaced as minister of veterans affairs by Hughes Lapointe. The plan set out the following:
> It is proposed to send to the next-of-kin a silver bar bearing the rank, name, regiment or corps, nature of casualty; this bar could be placed across the medals when they are mounted and framed. The cost of production would be in the neighbourhood of $60,000.[7]
The Memorial Bars were struck in sterling silver by Birks & Company of Montreal.[8] Each bar measured 67 mm by 17 mm. The width is slightly less than 2 36 mm medal ribbons. Most Canadians who served during the Second World War received at least 2 service medals: the Canadian Volunteer Service Medal and the War Medal, 1939–45. The corners of the bars were drilled with small holes to allow for the bar to be sewn onto the ribbons of the deceased's medals.
Each bar was engraved with four lines of text:
> Rank, forename initials, name, and postnominals of the individual commemorated
>
> Unit at time of death
>
> Died in His Country's Service / Died in Her Country's Service
>
> or
>
> Mort pour la patrie / Morte pour la patrie
>
> day-month-year of death
The criterion for the bar was quite simply "men and women who died as a result of war service."[9] This was to avoid criticism from the next of kin. By the end of 1951, most of the bars had been mailed. Each bar was sent in a brown envelope that contained a cream-coloured envelope bearing the Royal Arms of Canada on the reverse in purple. The bar was contained in a purple folder bearing the Royal Arms of Canada on the top in silver. No instructions were sent with the bar; it was simply assumed that family members would either affix the bar to the medal of their loved one or have it mounted in a frame. Given the fact that more than a decade had elapsed between when some of the individuals died and when the bars were engraved, approximately 8 percent were never delivered to the next of kin.
In the summer of 2008, a box of more than two thousand blank unnamed Memorial Bars was discovered in a government warehouse. These were transferred to the Department of National Defence where it was decided to revive the tradition of engraving the bars. The primary beneficiary of the estate of members of Her Majesty's Canadian Armed Forces who die as a result of an injury or disease related to military service since October 7, 2001, are in addition to the Memorial Cross presented with the Memorial Bar and a Memorial Scroll. Each of the post-2001 Memorial Bars is engraved in the same fashion as those from the Second World War. However, no regiment or unit is listed.
The Memorial Scroll
Memorial Scroll First World War issue.
Memorial Scroll modern issue.
Along with the reinstitution of the Memorial Bar, a Memorial Scroll was added to the memorial package that is presented to the next of kin. The scroll is based on that presented to the families of those who fell in the First World War. As with the Memorial Bar, only one scroll is presented to the primary beneficiary of the estate who, in theory, is also in possession of the deceased's medals. This is unlike the Memorial Cross for which up to three are now granted to recipients designated by the fallen Canadian Armed Forces member. The modern Canadian Memorial Scroll was introduced in January 2009.
The Memorial Scroll is printed on cream-coloured paper 21.5 cm by 27.9 cm with a thin black border displaying a large full-colour representation of the Royal Arms of Canada, the Queen's Royal Cypher (EIIR), and a modernized version of the First World War text. The gender of the text used in the scroll is altered as appropriate to the fallen member. Each scroll is contained in a black leatherette folder bearing the Royal Arms of Canada embossed on the cover in gold.
The Memorial Ribbon
Memorial Ribbon.
The Memorial Ribbon was instituted in 2012 to commemorate the death of a member of the Canadian Armed Forces who died as a result of an injury or illness directly attributable to military service sustained on or after October 1, 1947. Thus, the Memorial Ribbon is issued to the next of kin for all service-related deaths dating back to 1947. The first presentations were made in November 2012.
Origins: The Memorial Ribbon is based on the Memorial Cross that was instituted in 1919 to commemorate the death of personnel serving in the First World War. This memento is intended for family and close friends of a fallen Canadian Armed Forces member who are not in receipt of a Memorial Cross so that they might have an official and outward symbol of the personal loss that they have suffered. Consideration was given to increasing the number of Memorial Crosses presented per service-related death. However, the Memorial Ribbon was deemed a more fitting mechanism through which to acknowledge of a fallen Canadian Armed Forces member's broader circle of family and friends.
Criteria: Issued as a memento of personal loss of personnel of the Canadian Armed Forces who have laid down their lives for their country as a result of an injury or illness directly attributable to military service sustained on or after October 1, 1947. Up to five ribbons are issued to close loved ones who did not receive a Memorial Cross.
Insignia: A silver disc 15 mm in diameter with a raised edge of 1 mm displaying in the centre a Queen Elizabeth II issue Memorial Cross, the entire insignia pinned to a crossed length of purple ribbon. The reverse is plain.
Suspender: A single post pin soldered to the reverse of the insignia.
Naming: None.
Ribbon: A purple ribbon 6 cm in length folded through which the pin of the disc is placed.
Other: The purple ribbon is the same colour as the ribbon from which the original Memorial Crosses were issued with at the end of the Great War. The insignia is struck by the Royal Canadian Mint out of nickel-silver.
Number: 1,500.
The Royal Canadian Mounted Police Memorial Memento
Royal Canadian Mounted Police Memorial Memento.
The Royal Canadian Mounted Police had long sought a way to recognize the families of fallen RCMP members. The Canadian Memorial Cross that is presented to the families of members of the Canadian Armed Forces was seen as an excellent model to follow.
The Royal Canadian Mounted Police Memorial Memento was approved by RCMP Commissioner Juliano Zaccardelli in November 2005. The inaugural presentation took place on September 9, 2006, in Regina during the RCMP Memorial Service. In addition to the Memorial Memento, the names of those RCMP members killed in the performance of their duties continues to be entered into the RCMP Honour Roll Book and engraved on the Memorial Wall Plaques at the RCMP Depot in Regina and on the RCMP Headquarters Memorial in Ottawa.
Criteria: Presented to the next of kin of a member of the RCMP who has been killed in the performance of his or her duties. Three Memorial Mementos are presented to the next of kin, namely the mother, father, and spouse if alive. In the event that one of these recipients is deceased, then the Memorial Memento is presented to the eldest living child or the oldest brother or sister of the deceased.
Insignia: A circular silver medallion 34 mm in diameter. The circumference of the obverse displays thirteen natural maple leaves; the centre bears a police shield displaying the RCMP badge and a blank tablet at the base where the regimental number of the RCMP member is engraved.
Suspender: A straight bar suspender 36 mm wide bearing an inverted shield displaying a single maple leaf and another 30 mm wide bar bearing the words IN MEMORIAM. The bar suspended is attached to the disc of the medal by a pair of silver loops.
Naming: Engraved capitals, rank, initials surname, and service dates of the RCMP member commemorated.
Other: The RCMP Memorial Memento is struck by the Royal Canadian Mint in silver with an antique oxidized finish. The Memento is presented in a rectangular blue leatherette case bearing the arms of the RCMP in gold on the lid.
Number: 139 plus.
# 36
Commendations
The Canadian Armed Forces and a number of other organizations such as the lieutenant governors, territorial commissioners, RCMP, minister of veterans affairs, and St. John Ambulance all bestow commendations to recognize valuable or meritorious service. All of these awards come in the form of a small bar, and for the non-Canadian Armed Forces commendations there is also a lapel pin. Provisions have been made in the dress regulations of both the Canadian Armed Forces and the RCMP for these commendations to be worn on uniforms. All these commendations are ranked as "awards" and not "honours" as they are created and awarded by authorities other than the Sovereign. That is why, unlike the Mention in Dispatches and the old King's Commendation for Brave Conduct and the King's Commendation for Valuable Service in the Air (see chapter 5), insignia are not worn on or beside official undress ribbons and medals. Rather, they are worn separately on the breast pocket as a badge.
The Commander-in-Chief Unit Commendation and Canadian Forces Unit Commendation are roughly based on the Presidential Unit Citation awarded to members of the United States Army, Navy, and Air Force. While a battle honour is more akin to a campaign medal for a unit, the Commander-in-Chief Unit Commendation is similar in purpose to a Military Valour Decoration or Mention in Dispatches for a unit — the former in recognition of service, the latter in recognition of gallantry.
The American Presidential Unit Citation was created on February 26, 1942, and was originally styled the "Distinguished Unit Citation," but this was changed to "Presidential Unit Citation" on November 3, 1966. The principal requirement for the Presidential Unit Citation is extraordinary heroism in the presence of the enemy.
The Chief of the Defence Staff Commendation and the Command Commendation are similar in principle to the King's (later Queen's) Commendation for Brave Conduct, which was awarded to members of the Canadian Armed Forces until 1966, and the King's (later Queen's) Commendation for Valuable Service in the Air, which was awarded to members of the Royal Canadian Air Force (and other Commonwealth air forces) during the Second World War. The commendations were awarded "in recognition of meritorious service... or of gallantry not reaching the standard required for"[1] award of a gallantry medal. The Canadian Chief of the Defence Staff (CDS) Commendation and Command Commendations differ from the King's Commendation for Brave Conduct in that, like the King's Commendation for Valuable Service in the Air, the CDS Commendation can be awarded for meritorious service as well as bravery. The Command Commendation is more akin to a junior Mention in Dispatches. Both the CDS Commendation and Unit Commendation rank below a Mention in Dispatches.
The Commander-in-Chief Unit Commendation
Commander-in-Chief Unit Commendation.
Origins: The Commander-in-Chief Unit Commendation is roughly patterned after the Presidential Unit Citation used in the United States. It was established by the governor general on November 7, 2000, through a formal letter submitted to the chief of the defence staff. Recommendations are prepared and forwarded to the director of honours and recognition at National Defence Headquarters through the senior Canadian officer in the active theatre of operations. Awards are decided by the Canadian Forces' Decorations and Commendations Advisory Committee. The commendation is presented by the commander-in-chief.
Criteria: The commendation may be awarded to any unit or subunit of the Canadian Armed Forces or to any similar organization of a Commonwealth or foreign armed force working in conjunction with the Canadian Armed Forces that has performed an extraordinary deed or activity of a rare high standard in extremely hazardous circumstances. Commander-in-Chief Unit Commendations are restricted to war or warlike conditions in an active theatre of operations. Service that is considered beyond the demands of normal duty and that does not qualify for the Commander-in-Chief Unit Commendation may be considered for the Canadian Forces Unit Commendation.
Description: The commendation consists of a scroll, insignia, and pennant.
Scroll: A framed, gold-embossed scroll inscribed with the name of the formation, unit, or subunit being recognized. The scroll also bears the award citation and is signed by the commander-in-chief (the governor general).
Insignia: A gold bar bearing a vice-regal lion in the centre (the lion is in full colour). The insignia is worn for life by members of the unit or those attached to said unit during the action being recognized. Persons posted to a specific unit that has been recognized but who were not serving with the unit when the award was made wear the insignia while serving with that particular unit and cease to wear it when they leave. This insignia is worn by recipient members in the centre of the left breast pocket. Those who are only posted to a recipient unit wear it on the flap of the right breast pocket.
Pennant: A rectangular flag divided into three equal sections coloured navy blue, red, and sky blue (representing the three branches of the Canadian Armed Forces) and defaced in the centre with the vice-regal lion in full colour. The pennant is flown by the unit that has been recognized and can also be flown from the yardarm of HMC ships. The flying of the pennant shall not displace a commissioning or distinguishing pennant or flag; it may be flown at sea for special occasions approved by the senior officer present (such as Canada Day or the Queen's birthday); and it shall be flown in harbour from sunrise to sunset and when entering or leaving harbour.
Number: 7.
* • 1st Battalion of the Royal 22nd Regiment for its role in reopening the besieged Sarajevo Airport, July 1992.
* • 2nd Battalion of the Princess Patricia's Canadian Light Infantry for its display of courage during the 1993 Medak Pocket operation in the former Yugoslavia.
* • 3rd Battalion of the Princess Patricia's Canadian Light Infantry for spearheading Operation Apollo in Afghanistan, February-July 2002.
* • 1st Battalion of the Princess Patricia's Canadian Light Infantry for engagement in almost continuous combat in Afghanistan during twenty-nine major operations, July-August 2006.
* • 1st Battalion of the Royal Canadian Regiment for playing a decisive role in the largest ground combat operation in the history of NATO while deployed in Afghanistan, August-September 2006.
* • 3rd Battalion of the Royal 22nd Regiment for outstanding courage while battling insurgents in Afghanistan during a particularly difficult time for coalition forces, July 2007 to February 2008.
* • 1st Battalion of the 3rd Special Forces Group (U.S. Army) for heroism and outstanding combat ability in supporting a Canadian leg operation in Afghanistan, September 2006.
The Canadian Forces' Unit Commendation
Canadian Forces' Unit Commendation obverse.
Origins: A junior version of the Commander-in-Chief Unit Commendation, though the CF Unit Commendation predates the C-in-C Unit Commendation.
Criteria: Awarded to any formation, unit, or subunit of the Canadian Armed Forces or to any similar organization of a Commonwealth or foreign armed force working with or in conjunction with the Canadian Armed Forces that has performed a deed or activity considered beyond the demands of normal duty. It may be awarded to winners of competitions only under very exceptional circumstances.
Description: A framed, gold-embossed scroll inscribed with the name of the formation, unit, or subunit concerned. The scroll bears an appropriate citation and is signed by the chief of the defence staff.
Medallion: A table medal not intended for wear but meant to serve as a permanent symbol of the award. The medallion bears a modified version of the Canadian Armed Forces badge representing the three services.
Flag: A rectangular flag divided horizontally into three equal sections of navy blue, red, and sky blue and defaced in the centre with the symbols of the three services (an anchor, crossed swords, and an eagle) surrounded by a wreath of laurel leaves and surmounted by a Royal Crown. The flag is flown for only one year after the announcement of the award, then it is retained by the unit as a historical artifact.
The Chief of the Defence Staff Commendation
Chief of the Defence Staff Commendation.
Origins: Similar to the King's Commendation for Valuable Services in the Air, the CDS Commendation is a junior award intended to recognize valuable service that falls short of meriting bestowal of a Military Valour Decoration, Mention in Dispatches, bravery decoration, or Meritorious Service Decoration.
Criteria: Awarded to a member of the Canadian Armed Forces who has performed a deed or activity beyond the demands of normal duty. This award can also be presented to Commonwealth and foreign troops.
Insignia: A gold-coloured bar bearing three equally spaced maple leaves. The bar is worn centred on the left breast pocket.
Other: A recipient of the CDS Commendation also receives a framed, gold-embossed scroll inscribed with the member's name and an appropriate citation and signed by the CDS.
The Command Commendation
Command Commendation.
Canadian Army Commendation certificate.
Origins: A junior version of the Chief of the Defence Staff's Commendation.
Criteria: Awarded to a member of the Canadian Armed Forces who has performed a deed or activity affecting or reflecting well on the granting command. This award can also be presented to Commonwealth and foreign troops.
Insignia: A silver-coloured bar bearing three equally spaced maple leaves. The bar is worn centred on the left breast pocket.
Other: A recipient also receives a scroll signed by an appropriate National Defence Headquarters group principal or commander (sea, land, or air). Only eight commander or group principals have been authorized to award the Command Commendation.
Maximum Distribution Chart
(per 100 people under command over 6 months)
Conditions | Mention in Dispatches | CDS Commendation | Command Commendation
---|---|---|---
War | 1 | 2 | 3
Hostilities | 0.5 | 1 | 2
Field Operations | none | 0.5 | 1
Peacetime | none | few | few
The Vice-Regal and Commissioners' Commendation
Vice-Regal and Commissioners' Commendation.
Origins: While Canadians are eligible for admission or promotion in the Royal Victorian Order, many render important services to the lieutenant governors and territorial commissioners that though significant do not warrant appointment to the order. To remedy the need for a junior award to recognize these contributions, the Chancellery developed the Vice-Regal and Commissioners' Commendation.
In the early 1990s a proposal was developed to create a Vice-Regal Decoration — largely based on the Royal Victorian Order — to be awarded by lieutenant governors and territorial commissioners, but the plan was not approved because there was a general fear that such a decoration would not only complicate the Canadian honours system but would prompt various federal and provincial departments to insist on creating their own decorations and medals. The Queen approved use of the Crown on the insignia in July 1998, and the Commendation was created in January 1999.
Criteria: Awarded to persons who, in the course of their duties as paid or volunteer staff members, have performed one or more commendable acts that have benefited the office of a lieutenant governor or territorial commissioner. Nominations are submitted to the Chancellery by a lieutenant governor or territorial commissioner. Allotments are made for each lieutenant governor or territorial commissioner's term and are pro-rated on the size of their household (paid staff and volunteers). The Chancellery administers the program and maintains a register of recipients.
Insignia: A gold bar 35 mm wide and 10 mm high enamelled in blue. In the centre is a circle containing three red maple leaves surmounted by a Royal Crown. This bar is worn below official orders and medals on formal occasions. The lapel pin is a 10 mm circular badge depicting three red maple leaves in the centre, the circle surmounted by a Royal Crown. Recipients of the commendation also receive a certificate.
Other: The insignia was designed by the staff of the Canadian Heraldic Authority.
Vice-Regal and Commissioners' Commendation lapel pin.
Vice-Regal and Commissioners' Commendation certificate.
The Minister of Veterans Affairs Commendation
Minister of Veterans Affairs Commendation lapel pin and bar.
Origins: Veterans' groups had long encouraged the minister of veterans affairs to develop some sort of award that could be used to recognize veterans as well as non-veterans who have devoted their lives or a significant amount of time to assisting those who have served in Canada's armed forces. Although requests for such an award can be traced back to the 1960s, it was not until late 2000 that the concept for this award was fully developed.
Criteria: Awarded to those who have contributed to the care and well-being of veterans and/or those who have contributed to the remembrance of the contributions, sacrifices, and achievements of veterans. The commendation is intended primarily for veterans, but on rare occasions it may be awarded to non-veterans. There is no limit on the number of annual awards. Nominations are submitted to the minister of veterans affairs and reviewed by the Minister of Veterans Affairs Commendation Selection Committee. The Queen approved the use of the Royal Crown on the insignia in April 2001, and the commendation was officially created on March 4, 2002.
Insignia: A gold bar 35 mm wide and 10 mm high. At the centre is a stylized red poppy superimposed with a gold maple leaf and surmounted by a Royal Crown. This bar is worn below official orders and medals on formal occasions. The lapel pin is a 10 mm badge in the shape of a stylized red poppy edged in gold and superimposed with a gold maple leaf surmounted by a Royal Crown. Recipients of the commendation also receive a certificate.
Other: The insignia was designed by the staff at the Canadian Heraldic Authority.
Commendations of the Royal Canadian Mounted Police
As with the Canadian Armed Forces, the RCMP possesses a number of commendations for its uniformed and civilian members. The RCMP program is unique in that uniformed members receive a bar while civilians receive a lapel badge.
The Commissioner's Commendation
RCMP Commissioner's Commendations for first, second, and third awards.
Origins: The first RCMP Commendations were established in 1938 by Commissioner S.T. Wood and were originally only open to all RCMP members except commissioned officers. Two awards were established: the Commissioner's Commendation and the Commanding Officer's Commendation (later renamed the Command Commendation). Commissioner Leonard Nicholson broadened the program in 1957 to include commissioned officers, and since that time commendations have been awarded for bravery or outstanding service in an individual case or operation. In 1970 commendations were made available to civilian members of the force and members of the general public. Up to 1991 recipients of the Commissioner's Commendation received a plaque, while recipients of the Commanding Officer's Commendation received a letter. In 1991 specific Commendation Bars and Certificates were developed at the direction of Commissioner Norman Inkster. Members of the RCMP receive a Commendation Bar while civilians receive a lapel pin.
Criteria: Outstanding service in assistance rendered to a regular RCMP member performing his or her duty or law enforcement.
RCMP Commander's Commendations for first, second, and third awards.
Insignia: Force members receive a gilt bar 32 mm wide enamelled with blue and yellow and a buffalo head in the centre. Civilians receive a gold oval lapel pin 7 mm tall enamelled with yellow and blue and defaced in the centre by a buffalo head.
Other: A recipient is also presented with a certificate signed by the commissioner of the RCMP.
The Commander's Commendation
Criteria: Outstanding service in assistance rendered to a regular RCMP Member performing his or her duty or law enforcement at large.
Insignia: Force members receive a silver bar 32 mm wide enamelled with blue and yellow and set with a buffalo head in the centre. Civilians receive a silver oval lapel pin 7 mm tall enamelled yellow and blue and defaced in the centre by a buffalo head.
Other: This award was called the Commanding Officer's Commendation until 1997. Recipients are also presented with certificates signed by the division commander.
Commendations of the Most Venerable Order of the Hospital of St. John of Jerusalem
The Canadian Priory of the Order of St. John possesses two commendations: the Chancellor's Commendation and the Provincial Commendation. Both serve as an intermediate award between the Priory Vote of Thanks and appointment to the order. The concept behind these commendations is based on the Canadian Forces Chief of the Defence Staff and Command Commendations and the Vice-Regal Commendation.
The first awards of the commendation were made on June 23, 2006, when eight individuals were invested by Chancellor René Marin following the Order of St. John investiture dinner held in the West Block of Parliament. Unlike the Priory Vote of Thanks, these commendations may be awarded to those who are already members of the order.
The Chancellor's Commendation
Order of St. John Chancellor's Commendation.
Order of St. John Chancellor's Commendation lapel pin.
Order of St. John Chancellor's Commendation certificate.
Origins: Based on the Chief of the Defence Staff Commendation.
Criteria: The Chancellor's Commendation is "an exceptional distinction that is personally awarded by the Chancellor to recognize a particularly meritorious contribution on the national scene." The Chancellor's Commendation is awarded at the discretion of the Chancellor of the order.
Insignia: The insignia of the Chancellor's Commendation is a gold bar 3 cm wide bearing a white enamelled eight-pointed cross in the centre. A circular lapel badge gold in colour bearing a white enamelled eight-pointed cross is also included with the bar. The reverse is engraved with an issue number.
Other: The insignia was designed by Major Carl Gauthier. All Commendation Bars are presented with certificates signed by the Chancellor of the Order of St. John.
The Provincial and Territorial Council Commendation
Order of St. John Provincial/Territorial Council Commendation.
Order of St. John Provincial/Territorial Council Commendation lapel pin.
Origins: Based on the Canadian Forces Command Commendation.
Criteria: The Provincial Commendation is "an exceptional distinction that is awarded by the President of a Provincial Council to recognize a particularly meritorious contribution on the provincial scene." The Provincial Commendation is awarded on the recommendation of the local Honours and Awards Committee of St. John.
Insignia: The insignia is a silver bar 3 cm wide bearing a white enamelled eight-pointed cross in the centre. A circular lapel badge silver in colour bearing a white enamel St. John Cross is also included with the bar. The reverse is engraved with an issue number.
Other: The insignia was designed by Major Carl Gauthier. All Commendation Bars are presented with a certificate signed by the Chancellor of the Order of St. John or the president of a provincial council, depending on the level of award.
# 37
Components of Canadian Honours
Common Elements
All of Canada's honours have common elements incorporated into them. The maple leaf is used on almost every Canadian award, as is either the Royal Cypher and Royal Crown or an effigy of the Queen. These represent the two principal Canadian symbols: the maple leaf dates back to the 1840s, and its use by the St. John the Baptist Society and the Crown or Sovereign has been a constant in Canada for the past five hundred years. The use of these symbols and the fact that so many awards were designed by the same person creates the sense that though all Canadian awards extant were developed over a period of almost four decades they are all part of a continuum. As a result, the Canadian honours system with its diverse awards enjoys a remarkable continuity of design.
Breast Star
The only order in the Canadian honours system to be accompanied by a breast star is the Order of St. John (the top two levels: Bailiff/Dame Grand Cross and Knight/Dame of Justice or Grace). These insignia are worn on the lower left breast.
The Insignia
Insignia are the main parts of a particular order, decoration, or medal. They come in various shapes and sizes from snowflake or cross designs to decagonal or circular. Orders are generally enamelled, while decorations are in a shape (other than circular) such as a cross, star, or decagon. On medals the obverse usually depicts the Queen (who sanctions all awards), while the reverse bears some symbol or allegory associated with the purpose for the medal. The insignia of the various orders and valour or gallantry decorations are usually made of sterling silver, while other medals are made of bronze, cupronickel, or rhodium-plated metal.
Components of the Insignia
Suspender: This is the piece that links the insignia to the ribbon. In the case of honours worn around the neck (CV, CC, OC, CMM, and COM), the ribbon passes through a straight hanger connected to the insignia by a small ring. Other decorations have plain bars or stylized bars incorporating inverted fleurs-de-lys, a Crown, or a maple leaf.
Suspender Ring: This is a small ring mounted on top of a particular medal through which the ribbon passes.
Ribbon
The ribbon is the piece of material from which a particular insignia is suspended. Ribbons are often as symbolically rich as the actual medal or decoration. For instance, the ribbon for the Order of Canada is modelled on the Canadian flag, while the ribbon used on the Peacekeeping Service Medal symbolizes voluntarism (green), the colours of Canada (red and white), and the United Nations (light blue).
Bars
For valour, bravery, and Meritorious Service Decorations as well as long service medals, bars are awarded for subsequent awards or additional periods of service. These are sewn onto the ribbon. In the modern Canadian context, the term bar is also used for what in previous times would have been called a clasp, specifying the specific service being recognized on a campaign or service medal.
Clasps
Often also referred to as bars, clasps were awarded with the various operational service medals to denote service in a particular theatre or operation. Clasps were also awarded with the Queen's Medal for Champion Shot to denote the year in which the recipient won the competition. In the modern Canadian context, the only honour that still uses the term clasp is the Canadian Forces' Decoration.
Miniature Insignia
These are miniatures (usually half the size) of the full-size insignia, and are worn when the recipient is dressed in a mess dress or evening dress (gown, tuxedo, or tails).
Lapel Badges
A special aspect of Canadian orders and certain decorations is that the recipients are also authorized to wear a special lapel badge, usually a one-sixth-scale version of the actual award. This idea was introduced in 1967 when plans for the Order of Canada were being developed, though it would not be until 1972 that the lapel badges were actually instituted. The idea of lapel badges has its roots in several traditions. France has a long history of allowing recipients of various orders to wear a small swatch or boutonniere of ribbon to represent the highest award held by an individual. In the 1940s the United States began allowing recipients of decorations to wear a small lapel insignia made of metal — usually a miniature of the ribbon of the decoration in question. Canada appears to be the first to have initiated the practice of making the lapel badge in the form of the order itself and not the ribbon of the order. Recipients of the Order of Canada wear a small pin in the shape of a snowflake. The various levels are differentiated by the colour of the maple leaf in the centre of the lapel badge: red for Companions, gold for Officers, and silver for Members. Similarly, recipients of the Order of Military Merit and the Order of Merit of the Police Forces also receive a small pin in the shape of the cross of the order. As with the Order of Canada, the various levels are differentiated by the colour of the maple leaf in the centre of the lapel badge: red for Commanders, gold for Officers, and silver for Members. Recipients of all military valour, bravery, and Meritorious Service Decorations also receive lapel badges that are miniature versions of their insignia.
Undress Ribbons
Undress ribbon bar.
Undress ribbons are typically only worn by members of the Canadian Armed Forces, police, and firefighters, though members of various government agencies who wear official uniforms often wear undress ribbon bars when the actual medals are not worn. These are worn on the left breast, with the most senior award closest to the centre of the chest. The ribbons are worn in the order as set out in the Order of Precedence.
Undress Ribbon Bar Emblems
Several Canadian honours are accompanied by ribbon bar devices. For the Victoria Cross, Cross of Valour, Order of Military Merit, and Order of Merit of the Police Forces, a miniature version of the actual insignia is worn in the centre of the undress ribbon bar. A recipient of the Order of St. John, regardless of level, wears a silver Maltese cross in the centre of the undress ribbon. The Order of Canada is unique in that a recipient does not wear a miniature version of the insignia on the undress ribbon bar but rather a coloured maple leaf: red for Companions, gold for Officer, and silver for Members.
A recipient of a bar of the Victoria Cross or the Cross of Valour is authorized to wear two miniature versions of the particular decoration on the undress ribbon, signifying a second award. The addition of an insignia to denote a second award was originally the idea of King George V, who was concerned that recipients of multiple gallantry awards had no way to display their additional awards on their undress ribbons.[1] Thus, a small silver rose (rosette) was developed to be worn. In the Canadian context, this principle is personified by both rosettes (worn on the CD ribbon for multiple awards) and a variety of maple leaf devices that are worn on other decorations and medals to denote multiple awards or bars.
Silver Maple Leaf: Worn on the undress ribbon of the Member of the Order of Canada. Second awards of the Medal of Bravery and Meritorious Service Decorations are denoted by a silver maple leaf. Worn on the Special Service Medal to denote the award of two bars. Worn on the Gulf and Kuwait Medal and South-West Asia Medal to denote the award of a bar. Recipients of bars to the various Exemplary Service Medals wear one silver maple leaf for each bar. Also worn to denote receipt of a rotation bar on the General Campaign Stars, General Service Medals, and Operational Service Medals.
Gold Maple Leaf: Worn on the undress ribbon of the Officer of the Order of Canada, Medal of Military Valour, Star of Military Valour, and Star of Courage. Second awards of the MMV and SMV are denoted by a second gold maple leaf. Worn on the undress ribbon by recipients of a bar to the Star of Courage. Also worn to denote a third bar to the Special Service Medal. Worn to denote receipt of a two-rotation bar on the General
Campaign Stars, General Service Medals, and Operational Service Medals.
Red Maple Leaf: Worn on the undress ribbon of the Companion of the Order of Canada and to denote a fourth bar to the Special Service Medal. Also used to denote receipt of three-rotation bars on the General Campaign Stars, General Service Medals, and Operational Service Medals. When two red maple leaves are worn on an undress ribbon, it represents a five-maple-leaf rotation bar and a single-maple-leaf rotation bar, denoting six rotations. When a gold maple leaf is worn in conjunction with a red maple leaf, it denotes receipt of a single five-maple-leaf rotation bar, denoting five rotations. When a silver maple leaf is worn in conjunction with a red maple leaf, it denotes receipt of four rotation bars.
Silver Shield: Worn on the ribbon of the South-West Asia Service Medal to denote receipt of the AFGHANISTAN bar.
Oak Leaf Emblem: Worn to denote a Mention in Dispatches.
Silver Numerals: Worn on United Nations service medals to denote multiple tours.
Bronze Numerals: Worn on certain NATO service medals and the Multinational Observer Force Medal to denote multiple tours.
Square Bronze Numerals: Worn on various NATO service medals to denote multiple tours.
Silver Rosette: Worn to denote the award of each additional clasp to the Canadian Forces' Decoration.
Silver Maltese Cross: Worn on the undress ribbon of the Order of St. John and on the undress ribbon of the Service Medal of the Order of St. John. When worn on the undress ribbon of the Service Medal, it denotes periods of five years of voluntary service. Gold Maltese Crosses are worn to denote 10 years of service, while for 52 years of service a gold oak leaf is worn on the undress ribbon.
# 38
Canadian Honours in Heraldry
Heraldry — that is, the granting of arms — is in itself a form of honour and should not be overlooked. When an individual or corporate body petitions for arms and is awarded a grant, it signifies that the Crown has deemed that person or body worthy of recognition. A person or corporate body needs to demonstrate that he/she/it has made a positive contribution to the community.
The development of heraldry in Canada largely mirrors the development of our national honours system. At various times the French and British systems of heraldry have been used in Canada. (It is worth noting that English, Scottish, and Irish heraldry were traditionally administered separately and followed slightly different practices.) While Canadians acquired their own system of national honours in 1967, the development of a distinctive Canadian heraldic granting agency did not come until 1988 with the founding of the Canadian Heraldic Authority. Prior to that date, Canadians could petition the College of Arms in London or the Court of the Lord Lyon in Scotland to receive a grant of arms from one of the Crown's heraldic officers. To this day heraldic grants, like other honours, emanate from the Crown. Some notable Canadians who have been granted arms from these British authorities include Bruce Beatty, Alan Beddoe, R.B. Bennett, Samuel Bronfman, Sir George-Étienne Cartier, General Sir Arthur Currie, J.A.D. McCurdy, Sir Frederick Loomis, Jeanne Sauvé, Vincent Massey, Roland Michener, and others.
The Royal Arms of France were the first armorial bearings to be used in Canada and were typically surrounded by the Collars of the Order of the Holy Spirit and the Order of St. Michael, France's highest honours of the pre-revolutionary period. Similarly, the British Royal Arms were — and continue to be — surrounded by the "garter," the insignia of England's premier order, the Most Noble Order of the Garter, which contains the motto Honi soit qui mal y pense ("Evil be to him who evil thinks").
The practice of displaying the insignia of honours with coats of arms can be traced back to the early fifteenth century, shortly after the establishment of the Order of the Garter. Knights of the Garter would often encircle the shield of their arms with the motto of the order inscribed in gold on a blue garter with a gold buckle, and by the reign of Henry VII (1485–1509) this became common practice. It was only in the early nineteenth century that it became common practice for the insignia of orders to be hung from the shield of a person's coat of arms.
In more recent times it has been customary for the holders of the higher levels of the imperial orders of chivalry (Knights and Dames, Companions and Commanders) to surround their shield with the circlet bearing the motto of the order to which they belong. Only the motto circlet (and sometimes the collar itself) of the senior honour is used. Thus, a Knight of the Order of the British Empire who is also a Companion of the Order of the Bath surrounds his shield with the motto circlet of the Order of the British Empire because a KBE, as a higher level, takes precedence over a CB, even though the Order of the Bath is more senior in the British Order of Precedence by virtue of its date of foundation.
In the British system, Officers and Members of the various orders are not permitted to place a motto circlet around their shield. However, in the Canadian system all members of the three Canadian state orders (the Order of Canada, the Order of Military Merit, and the Order of Merit of the Police Forces) may do so regardless of their level of membership.
In Canada the general rule related to the inclusion of the insignia of orders, decorations, and medals in heraldic grants is that such insignia must be an official honour from the Crown. In the case of Commonwealth and foreign orders and decorations, the award must have been officially sanctioned by the Crown — that is, it must have been approved by the Government Honours Policy Subcommittee, which acts on behalf of the Queen.
In 1994 the Queen approved an augmentation to the Canadian coat of arms that saw the "motto circlet" of the Order of Canada placed around the shield. This follows a tradition, common in many countries, of incorporating some part of the insignia of the senior national honour in the national coat of arms.
Basic Heraldry
Armorial bearings consist of many different elements, and there are rules governing their general structure and just what can be included. The central part of any coat of arms is the "shield," sometimes called the "arms." Below the shield (and occasionally above the crest) is a "motto" on a scroll. The phrase or word placed here can be written in any language, though Latin, English, and French tend to be the most common in Canada.
Above the coat of arms is the "helm" or "helmet," and this is surrounded by what is known as "mantling." Above the helm is a "wreath" upon which sits the "crest." Crests were originally used as an identifying mark placed on top of a knight's helmet at a tournament. The helmet was shielded from the sun by colourful cloth mantling held in place by the wreath of twisted cloth. The crest is usually in the form of an animal, bird, or mythical beast. In the case of the Honourable J.J. Grant's arms, he chose a red lion rampant holding an abacus and a flag bearing the crest of the Fraser Highlanders.
As noted above, recipients of honours such as members of the Order of Military Merit are entitled to place a motto circlet around their shield. In the case of the Order of Military Merit, the motto circlet contains the motto of the order, while the motto circlet for the Order of Merit of the Police Forces simply contains CANADA MERIT MÉRITE.[1] Members of orders and holders of decorations or medals may also hang the insignia of these official honours from the bottom of their shields. A maximum of three insignia may be hung from the shield, the senior one in the centre.
Senior state figures such as lieutenant governors, privy councillors, and members at the highest level of the Order of Canada (Companions), the Order of Military Merit, the Order of Merit of the Police Forces, the Royal Victorian Order (Commanders), and the Order of St. John (Bailiffs or Dames Grand Cross) can also be granted what are known as "supporters." These are usually animals or mythical beasts placed on either side of the shield.
In heraldry the colouring of a coat of arms is done with "metals" and "colours." The metals are known as "or," that is, gold, and "argent," which is silver or white. The most common colours are: gules (red), azure (blue), sable (black), vert (green), and purpure (purple). One of the principal rules of colouring in heraldry is that metals cannot be placed on metals and colours cannot be placed on colours.
Pre-1987 Examples
Major-General Sir Frederick Loomis, one of Canada's great First World War military leaders, petitioned for arms in 1915 and was awarded a grant later that year. Loomis's postwar grant of 1920 displayed three of the honours he had been awarded by King George V during the war: Knight Commander of the Order of the Bath (his shield is surrounded by the motto of the order), Companion of the Order of St. Michael and St. George, and Companion of the Distinguished Service Order. The 1920 grant also included the arms of the cities of Cambrai and Mons as a tribute to the fact that these cities were liberated by the 3rd Canadian Division under his command in the final months of the First World War.
Vincent Massey petitioned the College of Arms for a grant of arms while he was Canadian Envoy Extraordinary and Minister Plenipotentiary in Washington, D.C. The grant came years before he was made a Companion of Honour or appointed governor general of Canada. Massey's bookplate from 1952 includes the insignia of the Order of the Companions of Honour and the Order of St. John. It is worth noting that, in addition to his receiving the Royal Victorian Chain, Massey was especially honoured by the Queen in 1963 by an augmentation to his personal arms of a canton (square in the upper left corner of the shield) bearing the Royal Crest of Canada. Although Massey was one of the first Canadians appointed to the Order of Canada, his death only a few months after being appointed to the order prevented a further alteration to his arms to include the insignia of the order he had played such an important role in creating.
Loomis Arms.
Massey Arms.
Pre-1987, Including the Order of Canada
As the first member of the Order of Canada, Roland Michener became the first person to have the insignia of the order included in his grant of arms. This was done in 1968 by the College of Arms in London. The Companion's insignia was hung from his shield, though the motto circlet of the order was not included.
Neither Jules Léger nor Edward Schreyer petitioned for arms during their terms as governor general. After 1987, Léger's family applied for and received a posthumous grant from the Canadian Heraldic Authority, a unique event. Similarly, Edward Schreyer would go on to petition for arms after his term as governor general had expired. Jeanne Sauvé was the last governor general to receive arms from the College of Arms in London.
Michener Arms.
Post-1988 Examples
After the creation of the Canadian Heraldic Authority, several new developments occurred relating to the inclusion of insignia of Canadian orders, decorations, and medals with grants of arms. Today the most common insignia to be included in grants of arms are the Canadian Forces' Decoration, the Order of St. John, the Order of Canada, the Order of Military Merit, the Royal Victorian Order, and the various provincial orders.
The Canadian Heraldic Authority will include in a painting accompanying a grant the insignia of an order, decoration, or medal included in the Canadian honours system. In the case of members of the Order of Canada, the Order of Military Merit, the Order of Merit of the Police Forces, the Royal Victorian Order, or the Order of St. John, there are specific guidelines for the display of insignia with coats of arms. Members of national orders are entitled to surround their shields with the motto circlet of the order, a privilege also used by Commanders of the Royal Victorian Order. At the highest two levels of the Order of St. John, members are allowed to place a large Maltese cross badge behind their shields.
The holders of the top levels of these orders (Companion in the case of the Order of Canada; Commander in the case of the Order of Military Merit, the Order of Merit of the Police Forces, and the Royal Victorian Order) are entitled to be granted supporters. This is also true of persons appointed to the top level of the Order of St. John of Jerusalem — a Bailiff or Dame Grand Cross.
The following section examines the inclusion of various insignia in grants of arms and is divided by the order, decoration, or medal received by the individual in question.
Order of Canada
General Jacques Dextraze, CC, CMM, CBE, DSO, CD
Dextraze Arms.
General Dextraze served in the Second World War and later rose to become chief of the defence staff. During the 1964 crisis in the Congo, he was made a Commander of the Order of the British Empire in recognition of his bravery. His grant includes the insignia of a Companion of the Order of Canada, Commander of the Order of the British Empire, and Commander of the Order of Military Merit.
Sylvia Olga Fedoruk, OC, SOM
Fedoruk Arms.
As lieutenant governor of Saskatchewan, Sylvia Fedoruk was entitled to supporters. Her grant includes the motto circlet of the Order of Canada, as well as the insignia of an Officer of the Order of Canada and that of the Saskatchewan Order of Merit, both on bows.
Joyce M. Bryant, CM, BEM
Bryant Arms.
Members of the Order of Canada are also entitled to use the motto of the order around their shield. Bryant's grant of arms includes her Member's insignia and that of the British Empire Medal, which she earned while serving with the Royal Canadian Air Force's Women's Division during the Second World War.
Order of Military Merit
Lieutenant-General James Cyrille Gervais, CMM, CD
Gervais Arms.
General Gervais's grant of arms includes only the insignia of the Order of Military Merit and the motto circlet.
Brigadier-General Owen William Lockyer, OMM, CD
Lockyer Arms.
William Lockyer's grant includes three insignia: those of an Officer of the Order of Military Merit, of an Officer of the Order of St. John, and the Canadian Forces' Decoration.
Royal Victorian Order and Royal Victorian Medal
Kevin Stewart MacLeod, CVO, CD
MacLeod Arms.
Kevin MacLeod's grant includes the insignia of a Commander of the Royal Victorian Order. The difference between this insignia and the more junior grade of Lieutenant of the Victorian Order is that it is hung from a neck ribbon and is not worn on a straight ribbon. As a Commander of the Order, MacLeod also has the motto circlet of the order around the shield and supporters, which in this case are a pair of bulls, the traditional symbol of the Clan MacLeod. It is interesting to note that this individual has been promoted through all three grades of the order that Canadians are eligible for: MVO, LVO, and CVO.
Yves Chevrier, CM, RVM
Chevrier Arms.
Yves Chevrier's grant is unique, since it is the only one granted by the Canadian Heraldic Authority that includes the Royal Victorian Medal. Chevrier is also a Member of the Order of Canada and was a long-serving member of the staff at Government House.
Order of St. John of Jerusalem
Eric Lawrence Barry, GCStJ, CD
Barry Arms.
As Bailiff Grand Cross of the Order of St. John, Eric Barry's grant includes the sash insignia of that grade. His shield is superimposed upon a Maltese cross (symbol of the Order), which is a further indication of his appointment to the most senior grade of the Order of St. John (Knights and Dames). Barry was also Lord Prior of the order.
Meritorious Service Decorations
Randolph Gherson, MSM
Gherson Arms.
Randolph Gherson's grant of arms includes the Meritorious Service Medal (Civil Division), which is hung from his shield.
Canadian Forces' Decoration
Philip Despard Pemberton Holmes, DFC, CD
Holmes Arms.
Philip Holmes's grant includes the Distinguished Flying Cross, which he earned as a pilot serving in the Royal Canadian Air Force during the Second World War, and Canadian Forces' Decoration and two bars, both hung from his shield.
James Harry MacKendrick, CD
MacKendrick Arms.
A former member of the Canadian Army and the Queen's Own Rifles of Canada, as well as a former deputy sergeant-at-arms of the Legislative Assembly of the Northwest Territories, James MacKendrick's grant includes the Canadian Forces' Decoration with three clasps (each represents ten years of additional service) hung from his shield.
Provincial Orders
Naranjan S. Dhalla
Dhalla Arms.
As a member of both the Order of Canada and the Order of Manitoba, Dhalla's grant includes both insignia. Because the Order of Canada takes precedence over the Order of British Columbia, it is hung on the dexter side.
Commonwealth Orders
The Right Honourable Pierre Trudeau, PC, CC, CH
Trudeau Arms.
Pierre Trudeau's grant includes the insignia of both the Order of Canada and the Order of the Companions of Honour. Trudeau is the only Canadian to have included the Order of the Companions of Honour in his grant.
Vera Roberts, OBE
Roberts Arms.
Although Canada ceased using the British honours system in 1967, Canadians who receive British and other Commonwealth honours approved by the Canadian government are permitted to include the insignia of their honours in their grants of arms. Vera Roberts's grant includes the insignia of an Officer of the Order of the British Empire hanging from her shield.
Foreign Orders
As with British and Commonwealth honours, when Canadians receive foreign honours that have been approved by the Government of Canada, they are permitted to include the insignia of their honours in their grants of arms. Canadian insignia are always given precedence over Commonwealth and foreign ones.
Roger Ouellette
Ouellette Arms.
Roger Ouellette's grant includes the insignia of two French honours: Chevalier of the Ordre national du Mérite and Chevalier of the Ordre des Palmes académiques.
John Donald Mitchell
Mitchell Arms.
As a Commander of the Order of St. John and a holder of the United States Silver Star (a gallantry award), John Mitchell's grant includes both insignia.
John Larkin Kerwin
Kerwin Arms.
The grant to John Larkin Kerwin includes both the insignia of a Companion of the Order of Canada and an Officer of France's Légion d'honneur.
With the relaxation of Canadian government policy toward the acceptance of Commonwealth and foreign honours, it is certain that the number of unusual honours included in Canadian grants will increase. Thus, a new and interesting field of study is certain to emerge in the years to come.
# 39
Self-Styled Orders, Unofficial Medals, and Other Elements Outside of the National Honours System
An official honour can only be created by the Sovereign, and this is true of all of Canada's national orders, decorations, and medals. In the case of ithe provinces, various honours have come into being after the granting of Royal Assent by the lieutenant governor, as personal representative of the Queen, to legislation passed by the assembly or through an order-in-council that the lieutenant governor, on behalf of the Crown, has assented to. These provincial honours are therefore deemed to emanate from the Crown in right of that particular province. Nevertheless, only when these provincial creations are incorporated into the national Order of Precedence for the wearing of orders, decorations, or medals can it be said that they have been officially recognized by the Queen in Right of Canada and can therefore be worn along with other official honours from the Crown.
Similarly, honours such as U.N. and NATO service medals that come from recognized head-of-state-level organizations that the Crown in Right of Canada has agreed to recognize are often incorporated into the national Order of Precedence so they can be worn with official Canadian honours.
These incorporations in the official order of wear, however, do not make these honours, either provincial or organizational, formally part of the Canadian honours system, nor does it implies that they awarded by or on behalf of the Queen. Only those honours that are created by and awarded on behalf of the Queen in Right of Canada are part of the Canadian honours system, while the Order of Precedence allows for the official wearing of a wider array of honours that are "recognized" by the Crown of Canada.
All honours that may be worn by Canadians in public are listed in the Order of Precedence that is maintained by the Chancellery of Honours at Rideau Hall. This list contains all officially sanctioned honours of the Crown — both provincial and federal. Any purported honour emanating from another source should never be worn alongside recognized national honours or worn in a way that implies official recognition — unless, of course, it is an approved Commonwealth or foreign honour (see chapter 34).
Some organizations and even municipalities have created their own "orders" to recognize achievement and have even made insignia and ribbons to imply they should be worn with other insignia. Unfortunately, this is not the case, and under no circumstances should such insignia be worn. Several provinces have unrecognized medals and decorations as well as recognized ones; again, the former are not permitted for wear. The Order of Precedence (see appendix 1) in this book is a clear guide to what may be worn in Canada: if something is not on the list, it should not be worn as a medal.
Municipalities and other organizations, notably the various protective services and police forces, would be well advised when considering the creation of an award to recognize its citizens or members to avoid confusion with recognized state honours. They should consider not using the words order, decoration, or medal, which have very specific meanings. The word award, on the other hand, creates no such problems. Also, these awards should not take the form of a wearable medal. Instead, one should opt for a plaque, a certificate, a trophy, or a table medallion.
There are also a number of fraternal or service organizations that use the term order in their titles but are not recognized national orders in the Canadian honours system. In many cases such organizations perform worthy charitable work. Nevertheless, the wearing of insignia related to such organizations should be confined to private events of those organizations and should never be combined with official honours or worn at public events such as Remembrance Day, state functions, formal dinners, and so forth. It is a distressing sight to attend an official event and see an excess of unofficial stars, medals, and other insignia that are unsanctioned and unrecognized by the state.
# 40
Protocol
Although Canada has one of the most egalitarian honours systems, it is nevertheless a system with its own hierarchy. To ensure that the appropriate irespect is accorded to each honour, they are to be worn in a specific order. This chapter sets out the various mounting styles, when to wear which medals, and how to care for your medals.
Why Should I Wear My Medals?
You have been officially recognized by the Crown for your service in one area or another. Your decorations are an outward display of not only your service but the honour that has been accorded to you by your country. You have been recognized in part to serve as an example to other citizens and wearing your medals is one way of serving as an example. Wearing your medals also raises the visibility of the Canadian honours system and provides an important service in this manner.
Why Should I Follow These Rules?
The rules for wearing orders, decorations, medals, and other awards exist to uniformly help you know what to wear and when to wear it. Unfortunately, there are some people who believe that "I earned this award therefore I can wear it any way I want." Sadly, this is a misinformed attitude that often leaves the person looking desperate to seem important.
By following these rules you are showing respect for the Crown and other recipients of Canadian honours. These rules are not unique to Canada. They are Canadian rules that have come to us through Commonwealth tradition and international custom.
When Should I Wear My Medals?
There are specific occasions when you may wear your orders, decorations, and medals. How you wear your medals is usually dictated by what order of dress you are wearing, whether it be a business suit, long evening dress, or dinner jacket (tuxedo). When you are invited to an event, the invitation will usually say what you should wear. Such notifications are usually included on the bottom of the invitation card. It will often say something along the lines of "Business Attire with Full-Size Decorations" or "Black Tie, Long Dress, with Miniatures (or Decorations)."
There is no rule preventing you from wearing your medals twenty-four hours a day, even on your pajamas. However, common sense should prevail. The invitation will usually indicate if medals are to be worn. When in doubt, just keep your medals in your pocket until you arrive at an event. Then, if you find it is an event at which medals are to be worn, you can pin them on.
Full-Size Medals or Miniature Medals?
The general rule to follow is that full-size medals are worn during the day, while miniature-size medals are worn during the evening.
Full-size medals are those officially presented to a person by the Crown. These are worn during the daytime, while miniature medals are typically worn during the evening with formal dress. Miniature medals are small replicas of the full-size insignia and are purchased privately.
Methods of Mounting
There are two methods of mounting used in this country: swing mounting and court mounting. Swing mounting quite literally allows the medals to swing around because the ribbons are attached to a straight bar. Court mounting is somewhat more complex. The medals are hung from their ribbons and then sewn onto a backing board to prevent them from flopping around. The board is concealed by the ribbons of the insignia they support. About 2 cm of the medal hangs below the backing board. Swing mounting used to be more practical, since medals required regular polishing, but it has largely been replaced by court mounting. Court mounting also prevents the medals from hitting one another and becoming damaged. In royal or vice-regal presence, court mounting is generally mandatory, as it is in the Canadian Armed Forces.
Court mounting originated in Germany in the mid-nineteenth century. The close links between the British and German Royal Households brought about the importation of this tradition to Britain and the Commonwealth. The advantage of court mounting is that the medals are firmly attached to a backing board, thus preventing them from coming in contact with one another. This limits the amount of wear incurred on the actual insignia, especially those with enamelled surfaces that are prone to chip or crack with frequent wear.
The court mounting rule for the Canadian Armed Forces is a maximum length of 10 cm from the top of the mounting to the bottom of the insignia. The RCMP has a slightly different length of 7 cm. For civilians either of these lengths is appropriate.
Dinner Jacket or Long Evening Dress
Men (Black Tie)
Men, black tie.
The miniature insignia of all orders, decorations, and medals should be worn suspended from a medal bar attached to the left lapel of the coat. Only one full-size neck badge should be worn suspended from a miniature-width ribbon. The ribbon is worn under the shirt collar over the tie so that the badge hangs 2.5 cm below the bow/knot. A miniature of this badge should be included in those worn on the medal bar. Only one star of an order should be worn on the left side of the coat.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Women (Long Evening Dress)
Women, long evening dress.
The miniature insignia of all orders, decorations, and medals should be worn suspended from a medal bar attached to the left side of the dress. If in possession of only one such award, it may be worn on a bow.
Only one full-size neck badge should be worn. This should be worn on a bow above the medal bar on the left shoulder. In place of a bow, women are permitted to wear one full-size neck badge around the neck on miniature-size ribbon. Only one star of an order should be worn on the left side of the dress.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Morning Dress
Men (Tailcoat or Director's Short Black Coat)
Men, morning dress.
Those full-size orders, decorations, and medals that are suspended from the medal bar are worn attached to the left side of the coat. Only one full-size neck badge should be worn suspended from a full-width ribbon. The ribbon is worn under the shirt collar so that the badge rests immediately below the knot of the tie.
Up to four stars of orders may be worn on the tailcoat, but only one star is worn on the director's short black coat, and it is attached to the left side below the insignia on the medal bar.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Women (Afternoon Dress)
Those full-size orders, decorations, and medals that are suspended from a medal bar are worn attached to the left side of the dress. If in possession of only one such award, it may be worn on a bow.
Only one full-size neck badge should be worn. This should be worn on a bow above the medal bar on the left shoulder. In place of a bow, women are permitted to wear one full-size neck badge around the neck on a full-size ribbon. Only one star of an order should be worn on the left side of the dress below all other insignia.
Women, afternoon dress.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Business Attire
Men (Business Suit)
Men, business suit and full-size medals.
Those full-size orders, decorations, and medals that are suspended from a medal bar are worn attached to the left side of the coat. Only one neck badge should be worn suspended from a full-width ribbon. The ribbon is worn under the shirt collar so that the badge rests on the tie immediately below the knot. The stars of orders should not be worn with business suits.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time.
Women (Dress or Suit)
Women, business suit or dress and full-size medals.
When attending a function where men are wearing business suits, women should wear their full-size orders, decorations, and medals that are worn from a medal bar. Only one full-size neck badge should be worn. This should be worn on a bow above the medal bar on the left shoulder or at the neck on a full-size ribbon. The stars of orders should not be worn on these occasions.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time.
Men, business suit with miniature medals.
Women, suit or dress with miniature medals.
NOTE: Miniatures may be worn on business attire for semi-formal evening events. On these occasions neck insignia are worn from miniature ribbon, unless being worn on a bow by a woman.
Uniforms
Since there are many different types of uniforms that can be worn, this section provides a general overview of how to wear your medals with various uniforms. Members of other uniformed organizations, such as the Canadian Armed Forces, police forces, St. John Ambulance, the Corps of Commissionaires, et cetera, should wear the insignia of their orders, decorations, and medals as laid down in the dress regulations of the organization to which they belong.
Open-Neck Tunic
(Worn when business attire or morning dress is called for)
Those full-size orders, decorations, and medals that are suspended from a medal bar are worn attached to the left side of the tunic. Miniatures are not worn with this order of dress.
Two neck badges may be worn suspended from a full-width ribbon. The senior neck decoration is worn at the neck, with the ribbon worn under the shirt collar so that the badge rests on the tie immediately below the knot. The second neck decoration is worn with the full-size ribbon protruding from the top button hole. Up to four stars of orders can be worn on the left breast of the tunic.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Mess Dress
(Worn when black tie/dinner jacket is worn)
The miniature insignia of all orders, decorations, and medals should be worn suspended from a medal bar attached to the left lapel of the mess dress jacket. Only one full-size neck badge should be worn suspended from a miniature-width ribbon. The ribbon is worn under the shirt collar so that the badge hangs 2.5 cm below the bow/knot. A miniature of this badge should be included in those worn on the medal bar. Up to four stars may be worn on the left side of the coat.
A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
High-Neck Tunic
(As worn by the RCMP, RNC, and a number of police services)
Those full-size orders, decorations, and medals that are suspended from a medal bar are worn attached to the left side of the tunic. Up to three neck badges may be worn suspended from a full-width ribbon. The senior neck decoration is worn at the neck. The ribbon is worn inside the collar of the tunic in such a way that the badge hangs out with approximately 2.5 cm of ribbon protruding from the collar. The second and third neck decorations are worn with the full-size ribbon protruding from the second and third button holes respectively.
Up to four stars of orders can be worn on the left breast of the tunic. Commendations are worn on the pocket as they are with an open-neck tunic. A person who has been awarded a commendation wears the full-size Commendation Bar 2.5 cm below the group of miniature medals. A maximum of four Commendation Bars may be worn at any given time. Commendations are not worn when wearing a breast star.
Grand Crosses of the Order of St. John
Bailiffs and Dames Grand Cross of the order are invested with a sash with a badge and a breast star. Depending on the occasion, the sash and breast star are worn in different manners.
Dinner Jacket or Long Evening Dress
Men
If you do not already have a decoration that is worn at the neck, it is permissible to wear the sash badge of the order around the neck since the sash is not worn with a black tie. The sash badge is detached from the sash and hung from a miniature ribbon. The ribbon is worn under the shirt collar so that the badge hangs 2.5 cm below the bow. A miniature of this badge should be included in those miniature medals worn on the medal bar. Only one full-size neck badge should be worn. The star is worn on the left side of the coat.
Women
If you do not already have a decoration worn at the neck, then you may wear the sash badge of the order on a full-size bow or around the neck on a miniature ribbon. The star is worn on the left side. A miniature of this badge should be included in those miniature medals worn on the medal bar.
Morning Dress
Men (Tailcoat or Short Director's Jacket)
A sash cannot be worn with morning dress. The star is worn on the left side. Up to four stars may be worn on a tailcoat, while only one star may be worn on a short director's jacket.
Women
The full sash is not worn by ladies for morning dress events. However, the breast star is attached to the left side.
Business Attire
Men and Women
No Grand Cross insignia is worn (i.e., not the sash, sash badge, or star).
Uniforms
The full sash is worn over the right shoulder on the outside of the uniform (under the shoulder strap where appropriate), with the badge of the order resting on the left hip. The breast star is attached to the centre of the pocket on the left breast. Up to four breast stars may be worn.
Mantles of the Order of St. John
In Canada the mantle of the order (sometimes referred to as robes) is usually only worn at investitures of the order and certain Priory events. They should not be worn unless expressly called for.
The robes can be worn over business attire, morning suits, long dresses, and uniforms. The robes cannot be worn over a dinner jacket or overcoat. You do not pin your medals to the robes; they should be worn underneath attached to the jacket or dress that you are wearing.
If wearing the robes along with the sopra vest, full-size or miniature medals are not worn. However, undress ribbons are worn on the sopra vest above the white cross in the middle of the chest.
Overcoats
Men and Woman
Overcoat with full-size medals.
Only those full-size orders, decorations, and medals mounted on a medal bar should be worn on an overcoat at outdoor functions such as Remembrance Day. Neither neck badges nor stars of orders should be worn (a neck badge may be worn underneath). Generally, orders, decorations, and medals should not be worn on an overcoat, because they are much more vulnerable to moisture and damage.
Lapel Badges
Men and Woman
Men, lapel pin.
Women, lapel pin.
Included in the insignia of some orders and decorations is a lapel badge or boutonniere. This badge may be worn on the left lapel of the coat with any order of dress at any time when not wearing full-size or miniature insignia or the undress ribbon. Women may wear this badge in a similar position on the dress or jacket. Only one such badge should be worn at a time.
Thus, if an individual is a member of the Order of St. John, has been awarded the Meritorious Service Medal, and has also been awarded a Chancellor or Provincial Council Commendation, he or she must choose which of the three lapel badges to wear at any given time.
Care and Cleaning
Swing-mounted full-size medals.
Court-mounted full-size medals.
Court-mounted miniature medals.
All modern Canadian operational service medals, Exemplary Service Medals, U.N./NATO service medals, and the Canadian Forces' Decoration are made of metals that do not require polishing. Indeed, they were designed to require no care other than proper handling. The Order of Canada, the Order of Military Merit, the Order of Merit of the Police Forces, the Royal Victorian Order, Canadian bravery and valour decorations, Meritorious Service Decorations, and the RCMP Long Service Medal are all made from silver or gilded silver and are plated or lacqured. These should only be cleaned with a silver dip cleaner, not an actual polish.
In the case of those honours that are enamelled — the Cross of Valour, Order of Canada, Order of Military Merit, Order of Merit of the Police Forces, and Royal Victorian Order — avoid using harsh cleaners and do not place them in an ultrasound cleaner (the type often used by jewellers to clean rings) because it may damage the enamel. Similarly, do not expose any enamelled insignia to extreme heat or cold; this, too, will crack the enamel. While quite beautiful and durable, these insignia should still be handled with care.
Postnominal Designations
Many Canadian orders and decorations are accompanied by postnominal designations or letters. These are sets of letters that recipients can place after their names to denote the fact that they have been honoured. In general one does not place periods between each letter in a set of postnominals, which should be placed in the following order:
* • Honours of the Crown (in accordance with the Order of Precedence) and privy councillorships (PC takes precedence over all honours, save the VC, GC, and CV).
* • Commonwealth and foreign honours.
* • Appointments in the following order:
* * Aide-de-Camp to the Queen (ADC).
* * Honorary Physician to the Queen (QHP).
* * Honorary Surgeon to the Queen (QHS).
* * Honorary Dental Surgeon to the Queen (QHDS).
* * Honorary Nursing Officer to the Queen (QHNO).
* * Honorary Chaplain to the Queen (QHC).
* * Aide-de-camp to the governor general or a lieutenant governor (AdeC).
* • Queen's Counsel (QC).
* • University degrees (BA, MA, MEd, MD, PhD, DPhil, et cetera; usually only the senior degree is used).
* • Religious orders and medical qualifications.
* • Fellowships of learned societies (with royal societies taking precedence over others; otherwise ranked in the order the societies were created).
* • Member of Parliament, Member of Provincial Parliament, Member of Legislative Assembly, Member of National Assembly (MP, MPP, MLA, MNA).
* • Membership in the Canadian Armed Forces (RCN, various branches of the Canadian Army where appropriate, RCAF).
# APPENDIX 1
The Canadian Order of Precedence of Honours, Awards, and Decorations
CANADIAN ORDERS, DECORATIONS, AND MEDALS DIRECTIVE, 1998
Extracted and adapted from P.C. 1998-591, April, 2, 1998
ORDER OF PRECEDENCE
1. 1. The sequence for wearing the insignia of Canadian orders, decorations, and medals, and the postnominal letters associated with the orders, decorations, and medals, are the following:
Victoria Cross (VC)
Cross of Valour (CV)
NATIONAL ORDERS
Order of Merit (OM)*
Companion of the Order of Canada (CC)
Officer of the Order of Canada (OC)
Member of the Order of Canada (CM)
Commander of the Order of Military Merit (CMM)
Commander of the Order of Merit of the Police Forces (COM)*
Commander of the Royal Victorian Order (CVO)
Officer of the Order of Military Merit (OMM)
Officer of the Order of Merit of the Police Forces (OOM)*
Lieutenant of the Royal Victorian Order (LVO)
Member of the Order of Military Merit (MMM)
Member of the Order of Merit of the Police Forces (MOM)*
Member of the Royal Victorian Order (MVO)
The Most Venerable Order of St. John of Jerusalem (all grades, postnominal letters only for internal use by the Order of St. John)
PROVINCIAL ORDERS
Ordre national du Québec (GOQ, OQ, CQ)
The Saskatchewan Order of Merit (SOM)
The Order of Ontario (OOnt)
The Order of British Columbia (OBC)
The Alberta Order of Excellence (AOE)
The Order of Prince Edward Island (OPEI)
Order of Manitoba (OM)*
Order of New Brunswick (ONB)*
Order of Nova Scotia (ONS)*
Order of Newfoundland (ONL)*
DECORATIONS
Star of Military Valour (SMV)
Star of Courage (SC)
Meritorious Service Cross (MSC)
Medal of Military Valour (MMV)
Medal of Bravery (MB)
Meritorious Service Medal (MSM)
Royal Victorian Medal (RVM)
MEDALS
Sacrifice Medal*
WAR AND OPERATIONAL SERVICE MEDALS
(See section 5 for complete list.)
Korea Medal
Canadian Volunteer Service Medal for Korea
Gulf and Kuwait Medal
Somalia Medal
South-West Asia Service Medal*
General Campaign Star*
* • Allied Force*
* • South-West Asia*
* • Expedition*
General Service Medal*
* • Allied Force*
* • South-West Asia*
* • Expedition*
Operational Service Medal*
* • South-West Asia*
* • Sierra Leone*
* • Haiti*
* • Sudan*
* • Humanitas*
* • Expedition*
Special Service Medal
* • Pakistan 1989–90
* • Alert
* • Humanitas
* • NATO/OTAN
* • Peace/Paix
* • Ranger*
* • Expedition*
Canadian Peacekeeping Service Medal*
UNITED NATIONS MEDALS
Service (Korea) (1950–54)
Emergency Force (Egypt/Sinai) (1956–67)
Truce Supervision Organization in Palestine (1948 to present) and Observer Group in Lebanon (1958)
Military Observation Group in India and Pakistan (1948 to present)
Operation in Congo (1960–64)
Temporary Executive Authority in West New Guinea (1962–63)
Yemen Observation Mission (1963–64)
Force in Cyprus (1964 to present)
India/Pakistan Observation Mission (1965–66)
Emergency Force Middle East (1973–79)
Disengagement Observation Force Golan Heights (1974 to present)
Interim Force in Lebanon (1978 to present)
Military Observation Group in Iran/Iraq (1988–91)
Transition Assistance Group (Namibia) (1989–90)
Observer Group in Central America (1989–92)
Iraq/Kuwait Observer Mission (1991 to present)
Angola Verification Mission (1988–97)
Mission for the Referendum in Western Sahara (1991 to present)
Observer Mission in El Salvador (1991–95)
Protection Force (Yugoslavia) (1992–95)
Advance Mission in Cambodia (1991–92)
Transitional Authority in Cambodia (1992–93)
Operation in Somalia (1992–93)
Operation in Mozambique (1992–94)
Observation Mission in Uganda/Rwanda (1993–94)
Assistance Mission in Rwanda (1993–96)
Mission in Haiti (1993 to present)
Mission in the Central African Republic (1998–2000)*
Preventive Deployment Force (Macedonia) (1995–99)*
Mission in Bosnia and Herzegovina (1995 to present)*
Mission of Observers in Prevlaka (Croatia) (1996 to present)*
Interim Administration Mission in Kosovo (1999 to present)*
Observer Mission in Sierra Leone (1999 to present)*
Mission in East Timor and Transitional Administration in East Timor (1999 to present)*
Mission in the Democratic Republic of the Congo (1999 to present)*
Mission in Ethiopia and Eritrea (2000 to present)*
Stabilization Mission in Haiti (2004 to present)*
Operation in Côte d'Ivoire (2004 to present)*
Mission in Sudan (2005 to present)*
Integrated Mission in Timor-Leste (2006 to present)*
Hybrid Mission with the African Union in Darfur (2007 to present)*
Mission in the Republic of South Sudan (2011 to present)*
Special Service (1995 to present)*
Headquarters*
NATO MEDALS
North Atlantic Treaty Organization (NATO) Medal for the Former Yugoslavia (1992-2002)*
NATO Medal for Kosovo (1999 to present)*
NATO Medal for the Former Yugoslav Republic of Macedonia (2001–02)*
Article 5 NATO Medal for Operation Eagle Assist (2001–02)*
Article 5 NATO Medal for Operation Active Endeavour (2001 to present)*
Non-Article 5 NATO Medal for Operations in the Balkans (2003 to present)*
Non-Article 5 NATO Medal for the NATO Training Mission in Iraq (2004 to present)*
Non-Article 5 NATO Medal for NATO Logistical Support to the African Union Mission in Sudan (2005 to present)*
Non-Article 5 NATO Medal for Service on Operations and Activities Approved by the North Atlantic Council in Relation to Africa (2008 to present)*
Non-Article 5 NATO Medal for Service on NATO Operation Unified Protector — Libya (2011)*
INTERNATIONAL COMMISSION AND ORGANIZATION MEDALS
International Commission for Supervision and Control (Indochina, 1954–73)
International Commission of Control and Supervision (Vietnam, 1973–75)
Multinational Force and Observers (Sinai, 1982 to present)
European Community Monitor Mission (Yugoslavia, 1991 to present)
International Force East Timor Service Medal (1999–2000)*
European Union Defence Policy Medal*
SPECIAL MEDALS
Polar Medal*
Sovereign's Volunteer Medal*
COMMEMORATIVE MEDALS
Canadian Centennial Medal (1967)
Queen Elizabeth II's Silver Jubilee Medal (1977)
125th Anniversary of the Confederation of Canada Medal (1992)
Queen Elizabeth II's Golden Jubilee Medal (2002)*
Queen Elizabeth II's Diamond Jubilee Medal (2012)*
LONG SERVICE AND GOOD CONDUCT MEDALS
RCMP Long Service Medal
Canadian Forces' Decoration (CD)
EXEMPLARY SERVICE MEDALS
Police Exemplary Service Medal
Corrections Exemplary Service Medal
Fire Services Exemplary Service Medal
Canadian Coast Guard Exemplary Service Medal
Emergency Medical Services Exemplary Service Medal*
Peace Officer Exemplary Service Medal*
SPECIAL MEDAL
Queen's Medal for Champion Shot
OTHER MEDALS
Ontario Medal for Good Citizenship (OMC)
Ontario Medal for Police Bravery
Ontario Medal for Firefighters Bravery
Saskatchewan Volunteer Medal (SVM)
Ontario Provincial Police Long Service and Good Conduct Medal
Service Medal of the Most Venerable Order of St. John of Jerusalem
Commissionaires Long Service Medal
Newfoundland and Labrador Award for Bravery*
Newfoundland and Labrador Volunteer Service Medal*
British Columbia Fire Service Long Service and Bravery Medals
Commemorative Medal for the Centennial of Saskatchewan (2005)
Alberta Centennial Medal (2005)
1. 2. The bar to the Special Service Medal is worn centred on the ribbon. If there is more than one bar, they are spaced evenly on the ribbon with the most recent uppermost. (This rule now also applies to the General Campaign Stars, General Service Medals, and Operational Service Medals.)*
1. 3. Commonwealth orders, decorations, and medals the award of which is approved by the Government of Canada are worn after the Canadian orders, decorations, and medals listed in section 1, precedence in each category being set by date of appointment or award.
1. 4. Foreign orders, decorations, and medals the award of which is approved by the Government of Canada are worn after the orders, decorations, and medals referred to in sections 1 and 3, precedence in each category being set by date of appointment or award.
1. 5. Notwithstanding sections 1, 3, and 4, a person who, prior to June 1, 1972, was a member of a British order or the recipient of a British decoration or medal referred to in this section, may wear the insignia of the decoration or medal together with the insignia of any Canadian order, decoration, or medal that the person is entitled to wear, the proper sequence being the following:
Victoria Cross (VC)
George Cross (GC)
Cross of Valour (CV)
Order of Merit (OM)
Order of the Companions of Honour (CH)
Companion of the Order of Canada (CC)
Officer of the Order of Canada (OC)
Member of the Order of Canada (CM)
Commander of the Order of Military Merit (CMM)
Companion of the Order of the Bath (CB)
Companion of the Order of St. Michael and St. George (CMG)
Commander of the Royal Victorian Order (CVO)
Commander of the Order of the British Empire (CBE)
Distinguished Service Order (DSO)
Officer of the Order of Military Merit (OMM)
Lieutenant of the Royal Victorian Order (LVO)
Officer of the Order of the British Empire (OBE)
Imperial Service Order (ISO)
Member of the Order of Military Merit (MMM)
Member of the Royal Victorian Order (MVO)
Member of the Order of the British Empire (MBE)
Member of the Royal Red Cross (RRC)
Distinguished Service Cross (DSC)
Military Cross (MC)
Distinguished Flying Cross (DFC)
Air Force Cross (AFC)
Star of Military Valour (SMV)
Star of Courage (SC)
Meritorious Service Cross (MSC)
Medal of Military Valour (MMV)
Medal of Bravery (MB)
Meritorious Service Medal (MSM)
Associate of the Royal Red Cross (ARRC)
The Most Venerable Order of St. John of Jerusalem (all grades, postnominal letters only for internal use by the Order of St. John of Jerusalem)
Provincial Orders (order of precedence as set out in section 1)
Distinguished Conduct Medal (DCM)
Conspicuous Gallantry Medal (CGM)
George Medal (GM)
Distinguished Service Medal (DSM)
Military Medal (MM)
Distinguished Flying Medal (DFM)
Air Force Medal (AFM)
Queen's Gallantry Medal (QGM)
Royal Victorian Medal (RVM)
British Empire Medal (BEM)
WAR AND OPERATIONAL SERVICE MEDALS
Africa General Service Medal (1902–56)
India General Service Medal (1908–35)
Naval General Service Medal (1915–62)
India General Service Medal (1936–39)
General Service Medal — Army and Air Force (1918–62)
General Service Medal (1962 to present)
1914 Star
1914–15 Star
British War Medal (1914–18)
Mercantile Marine War Medal (1914–18)
Victory Medal (1914–18)
Territorial Force War Medal (1914–19)
1939–45 Star
Atlantic Star
Arctic Star*
Air Crew Europe Star
Africa Star
Pacific Star
Burma Star
Italy Star
France and Germany Star
Defence Medal
Canadian Volunteer Service Medal
Newfoundland Second World War Volunteer Service Medal (see section 6)
War Medal (1939–45)
Korea Medal
Canadian Volunteer Service Medal for Korea
Gulf and Kuwait Medal
Somalia Medal
SPECIAL SERVICE MEDAL
(The order of precedence is as set out for the Special Service Medal with bars in section 1.)
UNITED NATIONS MEDALS
(The order of precedence is as set out for the United Nations Medals in section 1.)
INTERNATIONAL COMMISSION MEDALS
(The order of precedence is as set out for International Commission and Organization Medals in section 1.)
POLAR MEDALS
(The order of precedence is by order of date awarded.)
SPECIAL MEDALS
Polar Medal*
Sovereign's Volunteer Medal*
COMMEMORATIVE MEDALS
King George V's Silver Jubilee Medal (1935)
King George VI's Coronation Medal (1937)
Queen Elizabeth II's Coronation Medal (1953)
Canadian Centennial Medal (1967)
Queen Elizabeth II's Silver Jubilee Medal (1977)
125th Anniversary of the Confederation of Canada Medal (1992)
Queen Elizabeth II's Golden Jubilee Medal (2002)*
Queen Elizabeth II's Diamond Jubilee Medal (2012)*
LONG SERVICE AND GOOD CONDUCT MEDALS
Army Long Service and Good Conduct Medal
Naval Long Service and Good Conduct Medal
Air Force Long Service and Good Conduct Medal
RCMP Long Service Medal
Volunteer Officer's Decoration (VD)
Volunteer Long Service Medal
Colonial Auxiliary Forces Officer's Decoration (VD)
Colonial Auxiliary Forces Long Service Medal
Efficiency Decoration (ED)
Efficiency Medal
Naval Volunteer Reserve Decoration (VRD)
Naval Volunteer Reserve Long Service and Good Conduct Medal
Air Efficiency Award
Canadian Forces' Decoration (CD)
EXEMPLARY SERVICE MEDALS
(The order of precedence is as set out for Exemplary Service Medals in section 1.)
SPECIAL MEDAL
Queen's Medal for Champion Shot
OTHER MEDALS
(The order of precedence is as set out for Other Medals in section 1.)
1. 6. The Newfoundland Volunteer War Service Medal has the same precedence as the Canadian Volunteer Service Medal.
1. 7. The insignia of orders, decorations, and medals not listed in this directive, as well as foreign awards the award of which has not been approved by the Government of Canada, shall not be mounted or worn in conjunction with the orders, decorations, and medals listed in this directive.
1. 8. The insignia of orders, decorations, and medals shall not be worn by anyone other than the recipient of the orders, decorations, or medals.
*Denotes an award added to the Order of Precedence after passage of the 1998 order-in-council.
# APPENDIX 2
Instruments Creating and Amending Canadian Honours
This table below lists approval and order-in-council dates for the various elements of the Canadian honours system. The process for approving medals for wear by Canadians was not standardized until the late 1970s, and thus, some information is not complete. This is especially true with respect to the first twelve United Nations Service Medals issued to Canadians.
ORDER OF MERIT
Order-in-Council | Date of Registration | Change(s)
---|---|---
2010-1499 | November 26, 2010 | Return of the order to the Canadian Order of Precedence
ORDER OF CANADA
Order-in-Council | Date of Registration | Change(s)
---|---|---
1967-389 | March 2, 1967 | Establishment of the order
1972-809 | March 1, 1972 | Restructuring of the order to CC, OC, CM (abolition of Medal of Service and Medal of Courage)
1983-750 | March 17, 1983 | Enlargement of the order (OC and CM levels)
1994-2026 | December 6, 1994 | Increase in number of annual appointments of OC and CM
1995-694 | April 25, 1995 | Increase in number of annual appointments of CC
1997-552 | April 15, 1997 | Change in honorary appointment policy
1998-1373 | August 24, 1998 | Increase in number of annual appointments of OC and CM; also change in honorary appointment policy
1999-1743 | October 1, 1999 | Increase in number of annual appointments of OC and CM
2001-303 | March 1, 2001 | Increase in size of the Advisory Council
2013-413 | April 22, 2013 | Creation of an Extraordinary Division of the order for the governor general, vice-regal spouse, and the Royal Family
ORDER OF MILITARY MERIT
Order-in-Council | Date of Registration | Change(s)
---|---|---
1972-810 | May 1, 1972 | Establishment of the order
1972-1798 | August 24, 1972 | Modification of number of annual appointments
1977-2841 | October 6, 1977 | Modification of sections relating to honorary appointments
2003-1295 | July 21, 2003 | Administrative changes to text of letters patent and Constitution
2013-414 | April 22, 2013 | Creation of the Extraordinary Division of the order for the governor general and the Royal Family
2015-623 | May 25, 2015 | Modification to distribution of annual appointments
ORDER OF MERIT OF THE POLICE FORCES
Order-in-Council | Date of Registration | Change(s)
---|---|---
2000-1390 | August 24, 2000 | Establishment of the order
MILITARY VALOUR DECORATIONS
Order-in-Council | Date of Registration | Change(s)
---|---|---
1992-2577 | December 10, 1992 | Establishment of the VC, SMV, and MMV
DECORATIONS FOR BRAVERY
Order-in-Council | Date of Registration | Change(s)
---|---|---
1972-811 | May 1, 1972 | Establishment of the CV, SC, and MB
1997-123 | January 28, 1997 | Updating the membership of the Canadian Decorations Advisory Committee (expansion)
2005-1158 | June 9, 2005 | Modernization of regulations for Decorations for Bravery
2007-1889 | December 6, 2007 | To allow for nominations to be submitted up to two years after the findings of a quasi-judicial tribunal or coroner has reported on an incident or act of bravery
2015-1051 | June 30, 2015 | Updating the membership of the Canadian Decorations Advisor Committee (expansion)
MERITORIOUS SERVICE DECORATIONS
Order-in-Council | Date of Registration | Change(s)
---|---|---
1984-1831 | May 29, 1984 | Establishment of the MSC
1991-1060 | June 6, 1991 | Establishment of the Civil Division of the MSC and both Civil and Military Divisions of the MSM
1999-135 | February 4, 1999 | Expansion of the Civil Advisory Committee
2015-1050 | June 30, 2015 | Expansion of Civil Advisory Committee
MENTION IN DISPATCHES (MID)
Order-in-Council | Date of Registration | Change(s)
---|---|---
1991-861 | May 13, 1991 | Establishment of the MID regulations
SACRIFICE MEDAL
Order-in-Council | Date of Registration | Change(s)
---|---|---
2008-415 | February 28, 2008 | Establishment of the Sacrifice Medal
2009-1747 | October 16, 2009 | Amendment to allow all service-related deaths to be eligible for posthumous award of the medal.
WAR, OPERATIONAL, AND SPECIAL SERVICE MEDALS
Order-in-Council | Date of Registration | Change(s)
---|---|---
1984-1832 | May 29, 1984 | Establishment of the Special Service Medal
1992-2294 | November 12, 1992 | Re-establishment of the Special Service Medal
1991-1061 | June 6, 1991 | Establishment of the Pakistan 1989–90 bar to the SSM
1992-2437 | November 26, 1992 | Establishment of the Peace — Paix bar to the SSM
1992-2435 | Establishment of the Alert bar to the SSM
1992-2436 | Establishment of the NATO+OTAN bar to the SSM
1993-449 | March 9, 1993 | Establishment of the Humanitas bar to the SSM
1993-450 | Establishment of the Jugoslavija bar
1993-450 | Cancellation of Jugoslavija bar
1995-2003 | November 28, 1995 | Amendments to the criteria for the Alert bar
1995-2004 | Amendment to the criteria for the Peace — Paix bar
1995-2005 | Amendment to the criteria for the NATO+OTAN bar
1997-124 | January 28, 1997 | Amendment to the criteria for the Humanitas bar
1999-1742 | October 1, 1999 | Establishment of the Ranger bar to the SSM
1999-1741 | Regulations relating to approval of bars for the SSM
1991-860 | May 13, 1991 | Establishment of the Gulf and Kuwait Medal
1991-1177 | June 20, 1991 | Establishment of the Canadian Volunteer Service Medal for Korea
1996-1982 | December 19, 1996 | Establishment of the Somalia Medal
Act of Parliament 1997, Ch. 31 | April 25, 1997 | Establishment of the Canadian Peacekeeping Service Medal
1999-1858 | October 21, 1999 | Regulations for the Canadian Peacekeeping Service Medal
2002-1254 | July 17, 2002 | Establishment of the South-West Asia Medal
2004-287 | March 24, 2004 | Establishment of the General Campaign Star and the General Service Medal
2004-753 | June 17, 2004 | Establishment of the Allied Force bar for the General Campaign Star
2004-754 | Establishment of the Allied Force bar for the General Service Medal
2004-755 | Establishment of the ISAF bar for the General Campaign Star
2004-756 | Establishment of the ISAF bar for the General Service Medal
2006-810 | August 29, 2006 | Amendment to the criteria for the NATO bar to the SSM
2010-79 | January 14, 2010 | Repealing the 2004 Regulations for the General Campaign Star and General Service Medal to amend criteria and introduce campaign-specific ribbons for Allied Force and South-West Asia and the rotation bars
2010-80 | Repealing South-West Asia Service Medal regulations to amend criteria and introduce rotational bars
2010-248 | March 10, 2010 | Establishment of the General Campaign Star — Allied Force
2010-249 | Establishment of the General Campaign Star — Expedition
2010-250 | Establishment of the General Campaign Star — South-West Asia Service
2010-251 | Establishment of the General Service Medal — Allied Force
2010-252 | Establishment of the General Service Medal — Expedition
2010-253 | Establishment of the General Service Medal — South-West Asia
2010-704 | June 1, 2010 | Establishment of the Operational Service Medal
2010-1092 | September 2, 2010 | Cancellation of HUMANITAS bar to the SSM
2010-1093 | Establishment of the Operational Service Medal — South-West Asia
2010-1094 | Establishment of the Operational Service Medal — Sierra Leone
2010-1095 | Establishment of the Operational Service Medal — Haiti
2010-1096 | Establishment of the Operational Service Medal — Sudan
2010-1097 | Establishment of the Operational Service Medal — Humanitas
2010-1098 | Establishment of the Operational Service Medal — Expedition
2011-237 | February 14, 2011 | Amendment to regulations for South-West Asia Service Medal
2011-1380 | November 29, 2011 | Opening eligibility for Operational Service Medal to Government of Canada employees (non-CF)
2012-334 | March 20, 2012 | Amendment to Operational Service Medal regulations to correct French text
2014-606 | May 21, 2014 | Establishment of the Expeditions to the SSM
UNITED NATIONS SERVICE MEDALS1
Order-in-Council | Date of Registration | Incorporation in the Canadian Order of Precedence
---|---|---
* | ** | UNEF I (First U.N. Emergency Force)
* | ** | UNTSO (Truce Supervision Organization)
* | ** | UNOGIL (U.N. Observer Group in Lebanon)
* | ** | UNMOGIP (Observer Group in India and Pakistan)
* | ** | ONUC (Opération des Nations Unies au Congo)
* | ** | UNTEA (U.N. Temporary Executive Authority West New Guinea)
* | ** | UNYOM (U.N. Yemen Observer Mission)
* | ** | UNFICYP (U.N. Peacekeeping Force in Cyprus)
* | ** | UNIPOM (U.N. India-Pakistan Observer Mission)
* | ** | UNEF II (Second U.N. Emergency Force)
* | ** | UNDOF (U.N. Disengagement Observer Force)
* | ** | UNIFIL (U.N. Interim Force in Lebanon)
1989-1159 | June 15, 1989 | UNIIMOG (U.N. Iran-Iraq Military Observer Group)
1990-1370 | June 28, 1990 | UNTAG (U.N. Transition Assistance Group Namibia)
1990-1636 | July 27, 1990 | ONUCA (U.N. Observer Group in Central America)
1992-234 | February 6, 1992 | UNIKOM (U.N. Iraq-Kuwait Observer Mission)
1992-247-01 | March 27, 1992 | MINURSO (U.N. Mission for the Referendum in Western Sahara)
1992-260-01 | ONUSAL (U.N. Observer Mission in El Salvador)
1992-1736 | July 28, 1992 | UNPROFOR (U.N. Protection Force in Former Yugoslavia)
1993-1350 | June 16, 1993 | Extension of UNPROFOR Medal to Canadian Police
1993-1351 | UNAMIC (U.N. Advance Mission in Cambodia)
1992-1737 | July 28, 1992 | UNTAC (U.N. Transitional Authority in Cambodia)
1993-1352 | June 16, 1993 | Extension of UNTAC Medal to Canadian Police
1994-631 | April 21, 1994 | UNOSOM (U.N. Operation in Somalia)
1994-1152 | July 4, 1994 | ONUMOZ (U.N. Operation in Mozambique)
1995-1482 | August 20, 1995 | UNOMUR (U.N. Observer Mission in Uganda-Rwanda)
1995-1483 | August 30, 1995 | UNMIH (U.N. Mission in Haiti)
1998-53 | January 26, 1998 | MINUGUA (U.N. Mission for Verification of Human Rights in Guatemala)
1994-1153 | July 4, 1994 | UNAMIR (U.N. Assistance Mission for Rwanda)
* | ** | UNCRO (U.N. Confidence Restoration Operation in Croatia)
2000-278 | March 2, 2000 | UNPREDEP (U.N. Preventative Deployment Force, Former Yugoslavia)
1999-1859 | October 21, 1999 | MINURCA (U.N. Mission in the Central African Republic)
2000-278 | March 2, 2000 | U.N. Preventative Deployment Force (Macedonia)
2000-279 | UNMIBH (U.N. Mission in Bosnia and Herzegovina)
2000-280 | UNMOP (U.N. Mission of Observers in Prevlaka, Croatia)
2000-281 | UNMK (U.N. Mission in Kosovo)
2001-298 | March 1, 2001 | UNAMSIL (U.N. Mission in Sierra Leone)
2001-299 | UNAMET/UNTAET (U.N. Mission in East Timor and U.N. Transitional Administration in East Timor)
2001-300 | MONUC (U.N. Organization Mission in the Democratic Republic of the Congo)
2001-1405 | August 1, 2001 | UNMEE (U.N. Organization Mission in Ethiopia and Eritrea)
2005-1663 | September 26, 2005 | MINUSTAH (U.N. Stabilization Mission in Haiti)
2006-404 | May 18, 2006 | ONUCI (U.N. Operation in Côte d'Ivoire)
2007-385 | March 22, 2007 | UNMIS (U.N. Mission in Sudan)
2009-510 | April 2, 2009 | UNMIT (U.N. Integrated Mission in Timor-Leste)
2009-511 | UNAMID (African Union Hybrid Mission in Darfur)
2013-304 | March 19, 2013 | UNMISS (U.N. Mission in the Republic of South Sudan)
1999-2132 | December 1, 1999 | UNHQ (U.N. Headquarters Service)
1997-1216 | August 28, 1997 | UNSSM (U.N. Special Service)
1Listed in the order that the medals are worn (in Canada) and not by the date the medal was added to the Canadian honours system. Not all U.N. medals are listed here, since there are a number of missions in which Canada has not participated.
NATO MEDALS
Order-in-Council | Date of Registration | Incorporation in the Canadian Order of Precedence
---|---|---
1995-1484 | August 30, 1995 | NATO Medal, Former Yugoslavia bar
2000-282 | March 2, 2000 | NATO Medal, Kosovo bar
2002-1075 | June 13, 2002 | NATO Medal, Former Yugoslav Republic of Macedonia bar
2003-1298 | August 22, 2003 | Article 5 NATO Operation Eagle Assist Medal
2003-1297 | Article 5 NATO Operation Active Endeavour Medal
2003-1296 | Non-Article 5 NATO Operations in the Balkans Medal
2006-405 | August 29, 2006 | Non-Article 5 NATO Training Mission in Iraq
2007-384 | March 22, 2007 | Non-Article 5 NATO Logistical Support to African Union Mission in Sudan
2010-1394 | November 9, 2010 | Non-Article 5 NATO Medal for the North Atlantic Council Approved Operations and Activities in Relation to Africa
2012-951 | June 28, 2012 | Non-Article 5 NATO for Service on NATO Operation Unified Protector — Libya
INTERNATIONAL MISSION MEDALS
Order-in-Council | Date of Registration | Incorporation in the Canadian Order of Precedence
---|---|---
* | ** | International Commission for Supervision and Control Service
1973-2241 | July 27, 1973 | International Commission of Control and Supervision (Vietnam, 1973–75)
1986-1854 | August 13, 1986 | Multinational Force and Observers Medal
1994-629 | April 21, 1994 | European Community Monitor Mission Medal
2001-301 | March 14, 2001 | International Force East Timor Medal
2004-1552 | December 14, 2004 | European Security and Defence Policy Medal
CANADIAN FORCES' DECORATION
Order-in-Council | Date of Registration | Change(s)
---|---|---
1949-6335 | December 15, 1949 | Establishment of the Canadian Forces' Decoration following the King's approval
1950-3143 | June 14, 1950 | Making the insignia and ribbon for the CD the same for both the regular and reserve elements of the Canadian Armed Forces
1951-1000 | March 1, 1951 | To allow service in a Commonwealth navy, army, or air force to count toward the CD after five years of service in the Canadian Armed Forces
1951-6675 | December 19, 1951 | Allowed for members of the Canadian Armed Forces who had previous service in the auxiliary or reserve forces of a British Commonwealth force to be able to use that time toward earning the CD as members of a reserve component
1953-319 | March 5, 1953 | Allowing reservist serving in the regular force in Korea to count Korean War Service time toward the CD
1953-1869 | December 3, 1953 | New effigy and design, engraving of the medal on the rim, allowing for forfeiture of the CD and for the creation of regulations by the minister of national defence
1969-2019 | October 21, 1969 | To allow for officers serving as part of the Royal Canadian Sea Cadets to be made eligible for the CD
1976-2559 | October 21, 1976 | To remedy the difference in service required for split service in the reserve and regular components of the Canadian Armed Forces
1977-1748 | June 23, 1977 | Allowing for members of the Canadian Armed Forces who had previous service in the auxiliary or reserve force of a British Commonwealth force to use their previous service toward earning the CD (removal of requirement that reserve service in a Commonwealth force could only count toward service in the Reserve Force, and regular service in a Commonwealth force could only count toward service in the Regular Force)
1981-2310 | August 19, 1981 | Revoking the requirement that a member of the Regular Force who had transferred from the Reserve Force to serve five years in the Regular Force before qualifying
RCMP LONG SERVICE MEDAL
Order-in-Council | Date of Registration | Change(s)
---|---|---
1928-2169 | December 6, 1928 | Establishment of the RCMP Officer's Decoration and the RCMP Long Service Medal
1933-29 | January 14, 1934 | Cancellation of the RCMP Officer's Decoration and establishment of the RCMP Long Service Medal (for all ranks)
1945-1981 | March 16, 1945 | Allowance for service in the Canadian Armed Forces during the Second World War to count toward the Long Service Medal
1947-4877 | November 28, 1947 | Change of King George VI title to remove IND IMP and change to GEORGIVS VI DEI GRATIA REX
1951-4333 | August 15, 1951 | Amendments to medal regulations
1954-218 | February 12, 1954 | Establishment of the bronze, silver, and gold Long Service bars/clasps
1954-415 | March 18, 1954 | Rewording of Order-in-Council 1954-218
Amendment to Regulations | January 26, 2004 | Establishment of the forty-year clasp
Amendment to Regulations | January 30, 2012 | Establishment of the forty-five-year clasp
EXEMPLARY SERVICE MEDALS
Order-in-Council | Date of Registration | Change(s)
---|---|---
1983-2130 | July 14, 1983 | Establishment of the Police Exemplary Service Medal
1984-1830 | May 29, 1984 | Establishment of the Corrections Exemplary Service Medal
1985-2564 | August 13, 1985 | Establishment of the Fire Service Exemplary Service Medal
1990-2308 | 25 October 1990 | Establishment of the Canadian Coast Guard Exemplary Service Medal
1995-784 | May 16, 1995 | Amendments to the Corrections Exemplary Service Medal, broadening the definition of a corrections service officer
1994-632 | April 21, 1994 | Establishment of the Emergency Medical Services Exemplary Service Medal
2004-652 | May 11, 2004 | Establishment of the Peace Officer Exemplary Service Medal
2015-1049 | June 30, 2015 | Expansion of the definition of Peace Officer Exemplary Service Medal eligibility
COMMEMORATIVE MEDALS
Order-in-Council | Date of Registration | Change(s)
---|---|---
The Queen | June 27, 1966 | Establishment of the Canadian Centennial Medal
The Queen | May 18, 1977 | Establishment of the Queen Elizabeth II Silver Jubilee Medal
1992-962 | May 7, 1992 | Establishment of the 125th Anniversary of the Confederation of Canada Medal
2002-195 | February 15, 2002 | Establishment of the Queen Elizabeth II Golden Jubilee Medal
2011-1558 | December 8, 2011 | Establishment of the Queen Elizabeth II Diamond Jubilee Medal
POLAR MEDAL
Order-in-Council | Date of Registration | Change(s)
---|---|---
2015-796 | June 10, 2015 | Establishment of the Polar Medal
SOVEREIGN'S MEDAL FOR VOLUNTEERS
Order-in-Council | Date of Registration | Change(s)
---|---|---
2015-797 | June 10, 2015 | Establishment of the Sovereign's Medal for Volunteers
QUEEN'S MEDAL FOR CHAMPION SHOT
Order-in-Council | Date of Registration | Change(s)
---|---|---
1991-1282 | July 10, 1991 | Establishment of the Queen's Medal for Champion Shot
1994-636 | April 21, 1994 | Cessation of requirement to publish names of recipients in the Canada Gazette
MEMORIAL CROSS
Order-in-Council | Date of Registration | Change(s)
---|---|---
1919-2374 | December 1, 1919 | Establishment of the Memorial Cross
1940-4210 | August 27, 1940 | Revision of criteria for issue of the Memorial Cross to include airmen, merchant seamen, and firefighters
1950-5812 | December 5, 1950 | Criteria expanded to include those involved in U.N. missions and NATO operations
1966-2347 | December 15, 1966 | Augmentation to definition of widow and mother
1976-1715 | November 4, 1976 | Revision of criteria for issue of the Memorial Cross
2006-1449 | November 30, 2006 | Revision of criteria for issue of the Memorial Cross to allow for up to three Memorial Crosses to be presented to a relative, friend, or any other person
2008-1924 | December 12, 2008 | Revision of criteria to clarify rules for deaths before and after October 7, 2001
CANADIAN ARMED FORCES COMMENDATIONS
Order-in-Council | Date of Registration | Change(s)
---|---|---
Established by the governor general | November 7, 2000 | Establishment of the Commander-in-Chief Unit Commendation
Established by the chief of the defence staff | November 1980 | Establishment of the Canadian Forces Unit Commendation
Established by the chief of the defence staff | April 1974 | Establishment of the Chief of the Defence Staff Commendation
Established by the chief of the defence staff | July 1995 | Establishment of the Command Commendation
ORGANIZATIONAL LONG SERVICE MEDALS
Order-in-Council | Date of Registration | Change(s)
---|---|---
Established by the Corps of Commissionaires | August 20, 1948 | Establishment of the Canadian Corps of Commissionaires Long Service Medal
1998-301 | February 26, 1998 | Official recognition of the Canadian Corps of Commissionaires Medal
Notes
*No known order-in-council was issued to approve this medal; rather, it was internally approved by the Department of National Defence and subsequently included in the Canadian Order of Precedence.
**Date of authorization is unknown.
# APPENDIX 3
Equalization Chart: Pre-1967 Imperial Honours and Modern Canadian Honours
This chart provides a broad comparison between the imperial honours bestowed upon residents of Canada prior to 1967 and the modern Canadian honours system as it is constituted in 2014. It is difficult to compare any honour to such rarely bestowed accolades as the Order of the Garter or the Order of the Thistle; they remain unique honours for which no realistic comparison can be offered. Even in the case of appointments to the senior most levels of the Royal Victorian Order, Knight/Dame Grand Cross and Knight/Dame Commander, there is no equivalent in the modern Canadian honours system.
Modern Canadian Honour | Pre-1967 Imperial Honours
---|---
Victoria Cross | Victoria Cross
Cross of Valour | George Cross
Order of Merit | Order of Merit
Royal Victorian Chain | Royal Victorian Chain
Companion of the Order of Canada | Knight Grand Cross of the Order of the Bath (Civil Division), Knight Grand Cross of the Order of St. Michael and St. George, Knight Grand Cross of the Order of the British Empire (Civil Division), Order of the Companions of Honour
Officer of the Order of Canada | Knight Commander of the Order of the Bath, Knight Commander of the Order of St. Michael and St. George, Knight Commander of the Order of the British Empire
Member of the Order of Canada | Companion of the Order of the Bath, Companion of the Order of St. Michael and St. George, Commander of the Order of the British Empire
Commander of the Order of Military Merit | Knight Grand Cross/Knights Commander of the Order of the Bath (Military Division), Knight Grand Cross/Knight Commander of the Order of St. Michael and St. George, Knight Grand Cross/Knight Commander of the Order of the British Empire (Military Division)
Commander of the Order of Merit of the Police Forces | Knight Grand Cross/Knight Commander of the Order of the Bath (Civil Division), Knight Grand Cross/Knight Commander of the Order of St. Michael and St. George, Knight Grand Cross/Knight Commander of the Order of the British Empire (Civil Division)
Commander of the Royal Victorian Order | Commander of the Royal Victorian Order
Officer of the Order of Military Merit | Companion of the Order of the Bath (Military Division), Commander of the Order of the British Empire (Military Division)
Officer of Order of Merit of the Police Forces | Companion of the Order of the Bath (Civil Division), Commander of the Order of the British Empire (Civil Division)
Lieutenant of the Royal Victorian Order | Lieutenant of the Royal Victorian Order
Member of the Order of Military Merit | Officer of the Order of the British Empire (Military Division), Member of the Order of the British Empire (Military Division)
Member of the Order of Merit of the Police Forces | Officer of the Order of the British Empire (Civil Division), Member of the Order of the British Empire (Civil Division)
Member of the Royal Victorian Order | Member of the Royal Victorian Order
Order of St. John (all levels) | Order of St. John (all levels)
Star of Military Valour | Distinguished Service Order (gallantry), Distinguished Conduct Medal, Conspicuous Gallantry Medal (Naval and Air), Conspicuous Gallantry Cross*
Star of Courage | George Medal, Empire Gallantry Medal
Medal of Military Valour | Distinguished Service Cross, Military Cross, Distinguished Flying Cross, Distinguished Service Medal, Military Medal, Distinguished Flying Medal
Medal of Bravery | Order of the British Empire for Gallantry, Queen's Gallantry Medal,** Queen's Commendation for Brave Conduct
Meritorious Service Cross (military) | Distinguished Service Order (leadership), Air Force Cross, Air Force Medal
Meritorious Service Cross (civil) | Commander of the Order of the British Empire (Civil Division), Imperial Service Order
Meritorious Service Medal (military) | Officer of the Order of the British Empire (Military Division), Member of the Order of the British Empire (Military Division), Air Force Cross, Air Force Medal
Meritorious Service Medal (civil) | Officer of the Order of the British Empire (Civil Division), Member of the Order of the British Empire (Civil Division)
Royal Victorian Medal | Royal Victorian Medal
Sacrifice Medal | Wound stripe
Mention in Dispatches | Mention in Dispatches
*The Conspicuous Gallantry Cross was established in 1993 and replaced the Distinguished Service Order (for gallantry), Distinguished Conduct Medal, and Conspicuous Gallantry Medal and is very much the equivalent to Canada's Star of Military Valour.
**The Queen's Gallantry Medal was not established until 1974 and is largely based on the Medal of Bravery.
# APPENDIX 4
CANADIAN HONOURS SYSTEM ribbon chart
# NOTES
Chapter 1: Orders, Decorations, and Medals Around the World
1. Notably Ireland, Switzerland, and Uruguay. Sweden only bestows honours on members of its royal family and foreigners. China has no unified honours system, simply a broad array of military orders.
Chapter 2: Honours in Canada from the Earliest Times
1. Harold E. Gillingham, French Orders and Decorations (New York: American Numismatic Society, 1922), 18. The limits for the two top levels were increased to ten Grand Crosses and twenty-four Commanders in 1719, and to forty Grand Crosses and eighty Commanders in 1779.
Chapter 3: Imperial Honours in Canada from Confederation to the Centennial
1. Order-in-Council 1940-1430, May 9, 1940.
2. These numbers are presented here and have not been incorporated into the Canadian totals.
Chapter 4: Peerages, Baronetcies, and the Imperial Orders of Chivalry
1. Black was subsequently removed from the Queen's Privy Council for Canada and the Order of Canada.
2. Peter Galloway, The Order of the Bath (London: Philmore, 2006).
3. Royal Warrant constituting the Distinguished Service Order, London Gazette, November 9, 1886.
4. The Awards Coordination Committee was composed of the following officials: Ephrian H. Coleman (chairman), undersecretary of state; J.François Delaute (secretary), chief clerk for the Department of the Secretary of State; Sir Shuldham Redfern, secretary to the governor general; John Erskine Read, legal adviser, Department of External Affairs; Frederick L.C. Perierra, assistant secretary to the governor general; Charles Peter Edwards, lieutenant-commander, deputy minister of transport; H.F.G Letson, major-general, adjutant general; Duncan Kenneth MacTavish, commander, deputy judge advocate general, RCNVR; J.A. Sully, air vice-marshal, air member for personnel, RCAF; Robert Arthur Pennington, captain, RCN; E. Rollo Mainguy, captain, RCN; and Gustave Lanctot, Dominion archivist.
5. Library and Archives Canada (LAC), RG 2, vol. 377, minutes of the Awards Coordination Committee, July 12, 1943.
6. LAC, RG 2, vol. 377, H-5, arrangements for the establishment of the Canada Medal, 1943.
7. LAC, RG 2, vol. 377, minutes of the ACC, November 2, 1943. While the subcommittee reported on August 20, 1943, it was not until November 2, 1943, that its findings were entered into the minutes of the ACC. It is likely that the first report of the subcommittee related solely to the quota, while the others issued (ii–iv) were added at a later meeting of the subcommittee.
8. Toronto Globe, October 18, 1943.
9. LAC, RG 2, vol. 377, minutes of the Awards Coordination Committee, October 29, 1943.
10. LAC, MG 26 J5, Mackenzie King Diary, November 3, 1943.
11. LAC, RG 2, vol. 80, file H-5, 1945, Prime Minister's Office, "Statement re Canadian Civilian Honours and Awards," December 28, 1945.
12. LAC, MG 26 J5, Diary of Mackenzie King, March 6, 1946.
Chapter 5: Imperial Gallantry Decorations
1. Victoria Cross Royal Warrant, London Gazette, August 10, 1858.
2. There have been 1,354 individuals awarded the Victoria Cross, and three have gone on to receive a second award (bar).
3. P.E. Abbott and J.M.A. Tamplin, British Gallantry Awards (London: Nimrod Dix, 1981), 51.
4. Ibid., 78.
5. Edward Medal Royal Warrant, London Gazette, October 18, 1907.
6. Regulations for the King's Police Medal, 1909.
7. First listed in the London Gazette, May 9, 1843.
Chapter 6: Imperial Campaign Medals
1. For an excellent discussion of the origins and development of the North West Canada Medal, consult Barbara M. Wilson, The Military General Service, 1793–1814, Egypt 1882–89, North West Canada 1885 Medal Roll (London: Spink & Son, 1975).
2. LAC RG7 G 21, vol. 113, number 205, vol. 1.
3. LAC RG 24, file 5916, file HQ 51-4-64.
4. Ibid.
5. Taprell Dorling, Ribbons and Medals (London: George Philip & Son, 1963), 80.
Chapter 7: Imperial Long Service Decorations and Medals
1. Often these cases are found with a strip of paper glued to the bottom containing the name and rank of the recipient. These strips of paper were glued onto the box before the medals were sent to the engraver.
2. Times of London, June 26, 1920.
3. Between 1902, when the medal was instituted, and 1952, a total of 98,500 were awarded throughout the Commonwealth.
Chapter 8: Imperial Commemorative Medals
1. Nicholas Dirks, The Hollow Crown: Ethnohistory of an Indian Kingdom (Cambridge: Cambridge University Press, 1987), 129.
Chapter 10: The Confederation Medal
1. For a detailed examination of the Confederation Medal, see Darrel E. Kennedy, "The 1867 Confederation Medal: First Honour of the Dominion," Families (Ontario Genealogical Society) 42, no. 3 (August 2003).
2. LAC RG 2, series 1, vol. 8/R165-82-0-E, Order-in-Council 1868-598, June 3, 1868.
3. LAC RG 6, series A-1, vol. 180, file 753, Secretary of State.
Chapter 11: The Privy Councils
1. The three chief justices not sworn into the Privy Council of the United Kingdom were Sir William Richards (1875–79), Sir Richard Johnstone (1879–92), and the Honourable Patrick Kerwin (1954–63).
Chapter 13: The Order of Military Merit
1. The idea of a civilian and military division had been proposed by Vincent Massey in March 1966.
2. LAC R 5769, vol. 11, Lochnan Papers, Report of the Decorations Committee, January 9, 1970.
3. Ibid.
4. Gillingham, 20.
Chapter 15: The Queen's Personal Honours
1. Peter Galloway, David Stanley, and Stanley Martin, Royal Service Volume I (London: Stephen Austin and Sons Ltd., 1996), 40.
2. The reason for this appears to be because of the fact that the neck insignia worn by KCVOs and CVOs are identical.
3. In April 2004, Dix Noonan Webb sold a Collingwood CVO numbered 233 DUPLICATE.
Chapter 16: The Most Venerable Order of the Hospital of St. John of Jerusalem
1. Prior to 2003, Cleave and Company operated as J.R. Gaunt & Son.
Chapter 17: The Provincial Orders
1. Revised Statutes of Quebec, vol. 1, 1888, Chapter 7, 1601–07.
2. The Order of Ontario was established by an order-in-council.
3. Originally, the Advisory Council was chaired by the chief justice of Newfoundland and Labrador and included the clerk of the Executive Council, the president of Memorial University, and four other members appointed by the lieutenant governor-in-council (the provincial Cabinet).
Chapter 18: Military Valour Decorations
1. Royal Warrant constituting the Victoria Cross (amended), 1920.
2. M.J. Crook, The Evolution of the Victoria Cross (London: Midas Books, 1975), 30.
3. Queen's University Archives (QUA) Matheson Papers, File 296, DND proposal, October 27, 1966, Memorandum to the Defence Council, Awards for the Canadian Forces.
Chapter 19: Decorations for Bravery
1. Jack Boddington, "The Government of Canada Medal 'For Gallant and Humane Services Rendered in Life Saving from Shipwrecked Vessels,'" Orders and Medals Research Society Journal (Winter 1975): 199–203.
2. The George Crosses awarded were to Doreen Ashburnham-Ruffer (converted from an Albert Medal), Gordon Bastian (converted from an Albert Medal), Ernest Frost (converted from an Empire Gallantry Medal), Karl Gravell, Roderick Gray, James Hendry, John Patton, John Rennie, Arthur Ross, and Kenneth Spooner.
3. LAC MG 26 N4, Pearson Papers, vol. 1, 003-4, Memorandum for the Prime Minister, John Hodgson to Lester Pearson, March 19, 1968.
4. Cabinet Office Ceremonial Secretariat [COCS] C135, "Canadian Attitudes to Imperial Honours, 1943–1967," Esmond Butler to Sir Michael Adeane, April 20, 1972.
5. In particular, Colonel N.A. Buckingham and Commissioner Stevenson of the RCMP felt there should be separate awards that were not necessarily incorporated in the Order of Canada. Carl Lochnan at the Department of the Secretary of State seems to have been one of the first to suggest the creation of three separate awards.
6. LAC R5769, vol. 11, Lochnan Papers, Carl Lochnan to G.G.E. Steele, October 26, 1968.
7. This proposal was first devised on October 28, 1968, at the meeting of the subcommittee of the Government Decorations Committee.
8. Canada, House of Commons. Debates, December 30, 1971 (Ottawa: Queen's Printer, 1972), 10,859.
9. LAC RG 2, vol. 6318, Cabinet Minutes, July 29, 1971.
Chapter 20: The Meritorious Service Decorations
1. LAC RG 2, vol. 9, file H-5, Order-in-Council 1943-7964, Arrangements for the Creation of the Canada Medal, Section 7 (a)–(c), October 14, 1943.
2. LAC RG 2, vol. 377, Minutes of the ACC, October 15, 1943 (for Canada Medal as Canadian Meritorious Award), and DHH, 75/601, file 25, Memorandum from Major-General H.F.G. Letson to the Minister of National Defence (Army), March 29, 1944.
3. QUA, Matheson Papers, series 1, file 296, Proposal for the Establishment of the Forces Meritorious Decoration and Forces Meritorious Medal, October 27, 1966.
4. A single exception to this rule was made for members of the Canadian Armed Forces who rendered important services at the Canadian embassy in Tehran during the Iran Hostage Crisis in 1979–80.
5. Canadian Forces Honours Policy, September 15, 1982.
Chapter 22: The Sacrifice Medal
1. Grant Tyer, Drab Serge and Khaki Drill: The Foreign Service, Universal Service, Battle and Combat Dress Jackets of the Canadian Army, 1899–2003 (Ottawa: Parks Canada, 2004), 96.
2. Ibid., 163.
Chapter 24: The United Nations Service Medals
1. Lawrence Borts, United Nations Medals and Missions (Greenville: Keys Printing, 1998), 36.
2. John Blatherwick, Canadian Decorations and Medals (Toronto: Unitrade Press, 2003), 133.
Chapter 28: Commemorative Medals
1. LAC R 5769, vol. 12, file 16, Lochnan Papers.
2. LAC R 5769, vol. 12, file 16, Lochnan Papers, Centennial Commissioner to Maurice Lamontage, March 10, 1964.
3. LAC RG 2, Cabinet Submission from Paul Hellyer and Maurice Lamontage, May 11, 1965.
4. LAC R 5769, vol. 12, file 16, Lochnan Papers, O.G. Stoner (Privy Council Office Staff) to Jean Miquelon (Chairman of the Decorations Committee), September 30, 1965.
Chapter 30: Canadian Forces' Decoration
1. Despite the fact that the order-in-council requires the Canadian Forces' Decoration to be made from tombac, issues have been found in a variety of materials. After the Royal Canadian Mint ceased manufacturing the medals in the late 1980s, there have been a myriad of varieties and problems with the production.
2. Public Works and Government Service Canada [PWGSC] File on Manufacture of the CD, Memo from Director of Clothing General Engineering and Maintenance Claude Schreybert, July 15, 1993.
Chapter 32: Organizational Long Service Medals
1. The Service Medal was awarded for fifteen years of service to members of St. John Ambulance in Britain until 1990 when it was reduced to twelve years. The Service Medal has been awarded for twelve years of service in Australia, Canada, New Zealand, and South Africa since its inception. In certain colonies such as Ceylon (Sri Lanka), India, and tropical places, the Service Medal was awarded for ten years of service. This was because of the harsh living conditions experienced.
Chapter 33: Provincial Decorations and Medals
1. Records of recipients only kept since 1977.
2. Records of recipients only kept since 1978.
Chapter 35: The Memorial Cross and Next-of-Kin Mementos
1. Between 1948 and 1950 the Royal Canadian Mint struck 3,144,950 medals and 524,000 bars for the Canadian Volunteer Service Medal. LAC RG 120, vol. 16, Operations of Coining and Medals Department, Memorandum to the Master of the Mint, April 19, 1950.
2. Punches for the stars, 1939–45 War Medal, and Defence Medal were supplied by the Royal Mint for the production of dies. The punches and dies for the Canadian Volunteer Service Medal were made entirely by the Royal Canadian Mint.
3. LAC RG 120, vol. 16, Operations of Coining and Medals Department, Memorandum to the Master of the Mint, April 19, 1950.
4. Hansard, Debates of the House of Commons Canada, April 21, 1950, 1801.
5. Hansard, Debates of the House of Commons Canada, June 6, 1950, 3255.
6. Ibid.
7. LAC RG 2, series 18, vol. 137, C-2-5-1950 (Cabinet Documents #191-250), August 18, 1950.
8. It appears as though Birks was contracted to strike and name the bars mainly on account of the fact that the Royal Canadian Mint was overworked with the striking of new Canadian coinage and service medals.
9. Ibid.
Chapter 36: Commendations
1. J.M.A. Tamplin and P.E. Abbott, British Gallantry Awards (London: Doubleday Press, 1972), 343.
Chapter 37: Components of Canadian Honours
1. M.J. Crook, The Evolution of the Victoria Cross (London: Midas Books, 1975), 268.
Chapter 38: Canadian Honours in Heraldry
1. Prior to 2005, grants that included the motto circlet of the Order of Military Merit contained the text CANADA MERIT MÉRITE.
# BIBLIOGRAPHY
Abbott, P.E., and J.M.A. Tamplin. British Gallantry Awards. London: Nimrod Dix, 1981.
Abela, A.E. The Order of St. Michael and St. George in Malta and Maltese Knights of the British Realm. Valetta: Progress Publishing, 1988.
Alexander, E.G.M. South African Orders, Decorations and Medals. Cape Town: Human & Rousseau, 1986.
Barber, Richard. The Knight and Chivalry. London: Longman, 1970.
Bishop, Richard, John Hayward, and Diana Birch. British Battles and Medals. London: Spink & Son, 2006.
Blatherwick, John F. Canadian Orders, Decorations and Medals. Toronto: Unitrade Press, 2003.
Blondel, Jacques. Guide Pratique des Décorations. Paris: Lavauzelle, 1986.
Coddington, Deborah, and Alister Taylor. Honoured by the Queen: New Zealand. Auckland: Maxwell Printing, 1994.
Crook, M.J. The Evolution of the Victoria Cross. London: Midas Books, 1975.
Cullen, Tom. Maundy Gregory: Purveyor of Honours. London: Quality Book Club, 1975.
Dalzell, Mick, and Paul Riches. Mentioned in Despatches, 1948–1968. London: Privately Published, 1999.
de la Bere, Sir Ivan. The Queen's Orders of Chivalry. London: William Kimber and Company, 1961.
De-la-Noy, Michael. The Honours System. London: Allison & Busby, 1985.
Dorling, H. Taprell. Ribbons and Medals. London: George Phillip & Son, 1963.
Esden, Robert M. The Story of the Knights Bachelor. London: Imperial Society of Knights Bachelor, 1995.
Farwell, Byron. Queen Victoria's Little Wars. London: W.W. Norton Co, 1972.
Fauteux, Aegidius. Les Chevaliers de Saint-Louis en Canada. Montreal: Éditions des Dix, 1940.
Ford, Frank. "Titles of Honour in Canada." Queen's Quarterly 10, no. 1 (1902).
Fowler, T. Robert. Courage Rewarded: The Valour of Canadian Soliders Under Fire, 1900-2011. Ottawa: Valour Publications, 2013.
Gaffen, Fred. In the Eye of the Storm: A History of Canadian Peacekeeping. Toronto: Deneau & Wayne, 1987.
Galloway, Peter. The Most Ancient Order of the Thistle. London: Third Millennium Publishing, 2009.
____. The Most Honourable Order of the Bath. London: Third Millennium Publishing, 2006.
____. The Order of the British Empire. London: Austin and Son, 1996.
____. The Order of the Companions of Honour. London: Third Millennium Publishing, 2002.
____. The Order of St. Michael and St. George. London: Third Millennium Publishing, 2000.
____. The Order of Saint Patrick. Chichester: Phillmore & Company, 1983.
Galloway, Peter, et al. Royal Service: The Royal Victorian Order. London: Victorian Publishing, 1996.
Galloway, Strome. The White Cross in Canada, 1883–1983. Ottawa: St. John Priory of Canada, 1983.
Gardam, John. The Canadian Peacekeeper. Burnstown, ON: General Store Publishing House, 1992.
Gillingham, Harold. French Orders and Decorations. New York: The American Numismatic Society, 1922.
Glanfield, John. Bravest of the Brave: The Story of the Victoria Cross. Gloucestershire: Sutton Publishing, 2005.
Gordon, Lawrence. British Battles and Medals. Aldershot: Gale and Polden, 1947.
Haliday, Hugh. Valour Reconsidered: Inquiries into the Victoria Cross and Other Awards for Extreme Bravery. Toronto: Robin Brass Studio, 2006.
Hanham, Andrew. A Manual for the Wearing of Orders, Decorations and Medals. London: Spink & Son, 2005.
Heywood, Valentine. British Titles. London: Adam & Charles Black, 1951.
Johnson, Clive. Australians Awarded: A Comprehensive Reference for Military and Civilian Awards, Decorations, and Medals to Australians Since 1772. 2nd ed. Banksmeadow, Australia: Rennick Publications, 2014.
Joslin, Edward Charles. The Standard Catalogue of British Orders, Decorations and Medals. London: Spink & Son, 1976.
Joslin, Edward Charles, A.R. Litherland, and B.T. Simpkin. British Battles and Medals. London: Spink & Son, 1988.
Journals of the Orders and Medals Research Society, 1942–2013.
Lochnan, Carl. History of Honours in Canada. Unpublished Manuscript, 1976.
Mackenzie, Sir Edward. "The Baronets of Nova Scotia: Their Country and Cognizance," Proceedings and Transactions of the Royal Society of Canada 7, second series, May 1901.
Macmillan, Gerald. Honours for Sale: The Strange Story of Maundy Gregory. London: The Richards Press, 1954.
Mantle, Craig L., Paul Pellerin, Tom Douglas, Justin Wright, Mélanie Denis, eds. In Their Own Words: Canadian Stories of Valour and Bravery from Afghanistan, 2001–2007. Kingston: Canadian Defence Academy Press, 2013.
Maton, Michael. The National Honours and Awards of Australia. Kenthurst, Australia: Kangaroo Press, 1995.
Martin, Stanley. The Order of Merit: One Hundred Years of Matchless Honour. London: I.B. Tauris Press, 2007.
McCreery, Christopher. The Beginner's Guide to Canadian Honours. Toronto: Dundurn Press, 2008.
____. The Canadian Forces' Decoration. Ottawa: Department of National Defence, 2010.
____. The Canadian Honours System. Toronto: Dundurn Press, 2005.
____. Commemorative Medals of the Queen's Reign in Canada. Toronto: Dundurn Press, 2012.
____. "Honour, Nation and Citizenship in a Multicultural Polity: Federal Public Honours in Canada, 1917–1997." Doctoral Thesis, Queen's University, 2003.
____. Maintiens Le Droit: Recognizing Service, a History of the RCMP Long Service Medal. Ottawa: RCMP, 2014.
____. The Maple Leaf and the White Cross: A History of St. John Ambulance and the Most Venerable Order of St. John of Jerusalem in Canada. Toronto: Dundurn Press, 2008.
____. On Her Majesty's Service: Royal Honours and Recognition in Canada. Toronto: Dundurn Press, 2008.
____. The Order of Canada: Its Origins, History and Development. Toronto: University of Toronto Press, 2005.
____. The Order of Military Merit. Ottawa: Department of National Defence, 2012.
____. Preserving Prestige and Enhancing the Recognition of Excellence: A Review of the Canadian Honours System. Internal Report. Ottawa: Privy Council Office, 2011.
____. "Questions of Honour: Canadian Government Policy Towards British Titular Honours, 1867–1935." Master's Thesis, Queen's University, 1999.
McMillan, James. The Honours Game. London: Frewin Press, 1969.
Office of the Secretary to the Governor General. Guide for the Wearing of Orders, Decorations and Medals. Ottawa: Ottawa, 2013.
O'Shea, Phillip. Honours, Titles, Styles and Precedence in New Zealand. Wellington, New Zealand: Government Printer, 1977.
Patterson, Stephen. Royal Insignia: British and Foreign Orders of Chivalry from the Royal Collection. London: Merrell Holberton, 1998.
Report of the Committee on Honours and Rewards in War. London: HMSO, 1923.
Reynolds, Ken. Pro Valore: Canada's Victoria Cross. Ottawa: Directorate of History and Heritage, 2008.
Risk, James C. The History of the Order of the Bath and Its Insignia. London: Spink & Son, 1972.
Sainty, Guy Stair. The Orders of Saint John: The History, Structure, Membership and Modern Roles of the Five Hospitaller Orders of Saint John of Jerusalem. New York, 1991.
Sainty, Guy Stair, and Rafal Heydel Mankoo, eds. Burke's World Orders of Knighthood and Merit. London: Burke's Peerage, 2006.
Shell, Donald. The House of Lords. New York: Harvester Press, 1992.
Smyth, Sir John. The Victoria Cross, 1856–1964. London: Camelot Press, 1965.
Stacey, C.P. "Lord Monck and the Canadian Nation." Dalhousie Review 13 (1935).
Stone, Lawrence. "The Inflation of Honours." Past and Present 3 (1958).
Taylor, Grant. Drab Serge and Khaki Drill: The Foreign Service, Universal Service, Battle and Combat Dress Jackets of the Canadian Army, 1899–2003. Ottawa: Parks Canada, 2004.
Thomas, Wendy. The Register of Canadian Honours. Toronto: T.H. Best Printing Company, 1991.
Thomson, Donald Walter. "The Fate of Titles in Canada." Canadian Historical Review (1929).
Thyen, John R. Canada General Service Medal Roll, 1866–70. Winnipeg: Bunker to Bunker Books, 1998.
Tozer, Charles. The Insignia and Medals of the Grand Priory of the Most Venerable Order of the Hospital of St. John of Jerusalem. London: Hayward & Son, 1975.
Vickers, Hugo. Royal Orders: The Honours and the Honoured. London: Broadwall, 1996.
Walker, John. The Queen Has Been Pleased: The British Honours System at Work. London: Secker & Warburg, 1986.
Werlich, Robert. Orders and Decorations of All Nations. 2nd ed. Washington: Quaker Press, 1974.
Wigney, Edward H. Mentioned in Despatches of the CEF. Nepean, ON: Self-Published, 2000.
Wilson, Sir Arnold, and J.H.F. McEwen. Gallantry: Its Public Recognition and Reward in Peace and in War at Home and Abroad. London: Oxford University Press, 1939.
# PHOTO CREDITS
Author's Collection: 53, 72, 94, 113 (left), 125, 126, 129, 130, 147 (top), 151, 189, 190, 215, 222 (left), 223, 224, 226, 236, 238, 240, 252, 253, 254 (top), 276, 277, 278, 279, 282, 283(top), 288–290, 297, 300 (top; bottom left), 321 (top), 322 (bottom right), 325, 326, 333, 334, 342, 343 (bottom), 345–347, 351 (bottom), 377–379, 382 (right), 388, 390, 505 (left), 518 (top; second from bottom; bottom), 529 (bottom right; bottom left), 531 (left), 533 (left), 534 (bottom right), 536 (right), 538 (right), 542, 543, 554–556, 558 (bottom), 570, 580, 581, 590, 591 (top), 602 (top), 658, 690 (third ribbon, first row).
Assistant Commissioner Brian Brennan Collection: 540 (left).
British Medals by Rennie Alcock: 99, 100 (top), 140 (bottom), 319, 320.
Canadian War Museum, Tilston Memorial Collection of Canadian Military Medals: 89, 91 (bottom), 92, 93.
Department of Canadian Heritage: 6.
Department of National Defence, Directorate of Honours and Recognition: 27, 28, 40, 48, 50, 56 (bottom), 58, 59 (top), 63 (bottom left and right), 64, 68, 69, 77–79, 88, 90, 91 (top), 97, 98, 100 (bottom), 101, 102, 105–112, 113 (right), 114–117, 119, 128 (bottom), 142, 143, 146, 147 (bottom), 148–150, 153, 155–160, 162, 164, 167–176, 178, 180, 182–185, 191, 193, 195–199, 200 (right), 202, 203, 205, 206 (right), 208, 209, 211–214, 216–218, 220, 221, 222 (right), 227, 228, 230–235, 237, 274, 275, 280–282, 283 (bottom), 287, 296–299, 300 (bottom right), 302, 305, 307 (bottom), 308–10, 321 (bottom), 322 (top; bottom left), 324, 331 (bottom), 335, 337–341, 343 (top), 350, 351 (top), 352–357, 359–361, 363, 368, 369 (top; centre), 373–375, 382 (left), 383 (top; centre), 384–385, 386 (top; centre), 391–395, 397, 399–402, 405–411, 413, 414, 417–419, 421–424, 426–436, 438, 440–442, 445, 447, 448, 450–492, 495–504, 505 (right), 507–517, 518 (second from top), 520, 521, 528, 529 (top), 530, 531 (right), 533 (right), 534 (top; bottom left), 535, 536 (left), 546, 547, 550–552, 557, 558 (top left; top right), 559, 561–563, 565, 567–569, 571, 579, 597, 598–601, 602 (centre; bottom), 606, 607, 611, 614–622, 624, 626–629, 660, 661, 690 (except third ribbon, first row), 691, 692, 693 (except third and fourth ribbons, third row).
Dix Noonan Webb Ltd: 22, 95, 96, 121, 122, 123, 124, 131, 136, 137, 139, 140 (top), 187, 241–244.
Dr. Graham Neale Collection: 386 (bottom).
Glen Hodgins Collection: 128 (top).
Government of British Columbia: 583, 584.
Government of Newfoundland and Labrador: 181, 585.
Government of Saskatchewan: 578.
Library and Archives Canada: 32, 34, 37, 38, 53 (top), 54 (top), 55, 56 (top), 59 (bottom), 60 (bottom right), 61, 63 (top), 65 (left), 66, 67, 70, 71, 75, 76, 80, 82, 83, 254, (centre; bottom), 263, 265, 269, 270, 271, 273, 291, 292, 294, 315–16, 591 (bottom).
Lieutenant Colonel Jim Godefroy Collection: 135, 152.
Major Carl Gauthier Collection: 132, 200 (left), 596.
Mike Skific Collection: 206 (left).
Office of the Secretary to the Governor General, Chancellery of Honours: 284, 317, 369 (bottom), 522–525, 632, 634 (right), 635–644, 649–654, 659, 693 (third and fourth ribbons, third row).
Ontario Honours and Awards Secretariat: 575–577.
Queen's University Archives: 35, 54 (bottom), 60 (top left), 65 (right), 74, 81.
Royal Canadian Mounted Police: 307 (top), 538 (left), 539, 540 (right), 541, 608.
Royal Collection: 330, 331 (top).
Royal Military College of Canada: 634 (left).
Sam Kremer Collection (www.queensmedal.net): 245, 246, 249, 250.
Tanya Ursual, Military Antiquarian Inc: 229.
# COPYRIGHT
Copyright © Christopher McCreery, 2015
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise (except for brief passages for purpose of review) without the prior permission of Dundurn Press. Permission to photocopy should be requested from Access Copyright.
Editor: Michael Carroll
Designer: Courtney Horner
Front cover images: top — Chancellor of the Order of Canada insignia (Library and Archives Canada); bottom from left to right — Victoria Cross (Department of National Defence, Directorate of Honours and Recognition); General Campaign Star South-West Asia (Department of National Defence, Directorate of Honours and Recognition); Order of Merit, Queen Elizabeth II issue (Library and Archives Canada); Queen Elizabeth II's Diamond Jubilee Medal (Department of National Defence, Directorate of Honours and Recognition); Sacrifice Medal (Department of National Defence, Directorate of Honours and Recognition)
Back cover images: group photo credit — Sgt Ronald Duchesne, Rideau Hall, © Her Majesty the Queen in Right of Canada represented by the Office of the Secretary to the Governor General, 2015; Polar Medal credit — Dix Noonan Webb Ltd.
Author photo: © Paul Darrow
Library and Archives Canada Cataloguing in Publication
McCreery, Christopher, author
The Canadian honours system / Christopher McCreery ; foreword
by General John de Chastelain. -- second edition.
Includes bibliographical references and index.
Issued in print and electronic formats.
ISBN 978-1-4597-2415-0 (bound).--ISBN 978-1-4597-2416-7 (pdf).--
ISBN 978-1-4597-2417-4 (epub)
1. Decorations of honor--Canada. I. Title.
CR6257.M32 2015 929.8'171 C2015-904690-4
C2015-904691-2
We acknowledge the support of the **Canada Council for the Arts** and the **Ontario Arts Council** for our publishing program. We also acknowledge the financial support of the **Government of Canada** through the **Canada Book Fund** and **Livres Canada Books** , and the **Government of Ontario** through the **Ontario Book Publishing Tax Credit** and the **Ontario Media Development Corporation**.
Care has been taken to trace the ownership of copyright material used in this book. The author and the publisher welcome any information enabling them to rectify any references or credits in subsequent editions.
J. Kirk Howard, President
The publisher is not responsible for websites or their content unless they are owned by the publisher.
Visit us at: Dundurn.com | @dundurnpress | Facebook.com/dundurnpress | Pinterest.com/dundurnpress
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 3,777
|
\section{Introduction}
Quantum walks (QWs) are considered to be a quantum analog of classical random walks.
The system and the dynamics of QWs have some similarities to those of random walks, but the behavior of QWs is different from that of random walks in terms of their probability distributions .
In general, the behavior of the QWs can not be predicted based on our intuition.
A 3-period time-dependent QW which we are going to consider in this paper leads to an interesting behavior.
We study this behavior after a large number of discrete time steps and describe it as a long-time limit theorem.
The theorem will be given as a convergence in distribution on a rescaled space by time. The fact that the relevant scale is time itself and not its square root has been observed from the very first papers in the subject, \cite{AharonovDavidovichZagury1993}.
For a time-independent standard QW on the line, a limit distribution was obtained by Konno~\cite{Konno2002a,Konno2005} in 2002 for the first time and the limit density function has a representation similar to an arcsine law, in marked contrast to a Gauss distribution which appears for classical random walks under appropriate conditions.
Time-dependent QWs were numerically studied in some papers~\cite{MackayBartlettStephensonSanders2002,RibeiroMilmanMosseri2004,BanulsNavarretePerezRoldanSoriano2006,Romanelli2009} and some limit theorems were analytically derived \cite{MachidaKonno2010,Machida2011,Machida2013b,IdeKonnoMachidaSegawa2011}.
In particular, Machida and Konno~\cite{MachidaKonno2010} treated a 2-period discrete-time QW on the line whose time evolution is given by two unitary matrices which are used as coin-flip operators.
The long time behavior of the 2-period time-dependent walk can be completely determined by one of the two matrices according to the determinant of the product of both of them.
In this paper we define a 3-period time-dependent discrete-time QW on the line and we will see that this 3-period time-dependent walk also exhibits interesting behavior.
The motivation for the analytical study for the 3-period time-dependent walk done here comes from numerical studies done in Ribeiro et al.~\cite{RibeiroMilmanMosseri2004}.
Besides periodic time-dependent walks, they also looked at time-dependent QWs whose coin-flip operator was controlled by a quasiperiodic sequence or a random sequence.
According to their result, we can expect that the long time behavior of a walk with a long period is sub-ballistic or diffusive.
That means that as the length of the period increases, the behavior of the periodic time-dependent walks gets either less ballistic or more diffusive departing form the behavior of a time-independent quantum walk.
So, we would see a different behavior for a periodic QW depending on the length of the period, and this would be important in order to discuss the relationship between QWs and random walks.
We will define a 3-period time-dependent QW on the line in the following section.
The walker starts from the origin on the lattice $\mathbb{Z}=\left\{0,\pm 1,\pm2,\ldots\right\}$ at time 0 and from its state at time $t\in\left\{0,1,2,\ldots\right\}$ one gets the state at time $t+1$ after operating with a coin-flip operator and a position-shift operator.
In our model the coin operator is 3-periodic as a function of time $t$, and we use just one and the same coin-flip operator in the evolution.
For the 3-period time-dependent walk, we give a limit theorem as $t\to\infty$ in Sec.~\ref{sec:limit_th}.
The proof of the theorem is based on Fourier analysis and is included in the same section.
In the final section, we give a summary and a discussion of our result.
There are two appendices: in the first one we show how the analytical proof can be made to work in the case of some unitary (as opposed to orthogonal) operators.
In the second one we look at a number of models not covered by our analytical results and give some interesting numerical evidence of their limiting behavior.
\section{Definition of a 3-period time-dependent QW on the line}
\label{sec:definition}
In this paper we deal with a discrete-time 2-state QW on the line and we give a 3-periodic time evolution rule for the walk.
The total system of a discrete-time 2-state QWs on the line is defined in a tensor space $\mathcal{H}_p\otimes\mathcal{H}_c$, where $\mathcal{H}_p$ is called a position Hilbert space which is spanned by an orthogonal normalized basis $\left\{\ket{x}:\,x\in\mathbb{Z}\right\}$ and $\mathcal{H}_c$ is called a coin Hilbert space which is spanned by an orthogonal normalized basis $\left\{\ket{0},\ket{1}\right\}$.
Let $\ket{\psi_{t}(x)} \in \mathcal{H}_c$ be the state of the walker at position $x$ at time $t$.
The state of the 2-state QW on the line at time $t$ is expressed by $\ket{\Psi_t}=\sum_{x\in\mathbb{Z}}\ket{x}\otimes\ket{\psi_{t}(x)}\in\mathcal{H}_p\otimes\mathcal{H}_c$.
In particular, we focus on a 3-period time-dependent discrete-time QW whose coin-state is given by
\begin{align}
C=&\cos\theta\ket{0}\bra{0}+\sin\theta\ket{0}\bra{1}+\sin\theta\ket{1}\bra{0}-\cos\theta\ket{1}\bra{1}\nonumber\\
=&c\ket{0}\bra{0}+s\ket{0}\bra{1}+s\ket{1}\bra{0}-c\ket{1}\bra{1},
\label{eq:coin-flip operator}
\end{align}
with $\theta\in [0,2\pi)$ and we have abbreviated $\cos\theta, \sin\theta$ to $c, s$ in Eq.~(\ref{eq:coin-flip operator}).
The total system at time $t$ evolves to the next state at time $t+1$ according to the time evolution rule
\begin{equation}
\ket{\Psi_{t+1}}=\left\{\begin{array}{ll}
\tilde{S}\tilde{C}\ket{\Psi_t}& (t=0,1 \mod 3)\\[1mm]
\tilde{S}\ket{\Psi_t}& (t=2 \mod 3)
\end{array}\right.,
\label{eq:time-evolution}
\end{equation}
where
\begin{align}
\tilde{C}=&\sum_{x\in\mathbb{Z}}\ket{x}\bra{x}\otimes C,\\
\tilde{S}=&\sum_{x\in\mathbb{Z}}\ket{x-1}\bra{x}\otimes\ket{0}\bra{0}+\ket{x+1}\bra{x}\otimes\ket{1}\bra{1}.
\end{align}
The time evolution of the state $\ket{\Psi_t}$ depends on the value $t \mod 3$.
Equation~(\ref{eq:time-evolution}) states that the position of the walker gets shifted after the coin-flip operation has been completed at time $t=0,1 \mod 3$, and it just gets shifted without any coin-flip operation at time $t=2 \mod 3$.
Here, we don't take $\theta=0,\frac{\pi}{2},\pi,\frac{3\pi}{2}$ because the behavior of the walker would be trivial.
Under the condition $\braket{\Psi_0|\Psi_0}=1$, the quantum walker can be observed at position $x$ at time $t$ with probability
\begin{equation}
\mathbb{P}(X_t=x)=\bra{\Psi_t}\biggl\{\ket{x}\bra{x}\otimes (\ket{0}\bra{0}+\ket{1}\bra{1})\biggr\}\ket{\Psi_t},
\end{equation}
where $X_t$ is a random variable and denotes the position of the walker at time $t$, regardless of the spin orientation.
The probability distribution evolves as a function of time $t$, as numerically shown in Fig~\ref{fig:time-probability}.
Actually, this linear behavior is reflected in a limit theorem which will show up after this section.
We also show how the time evolution of the probability distribution depends on the parameter $\theta$ of the coin-flip operator $C$ in Fig.~\ref{fig:theta-probability}.
We will analyze the long time behavior of this probability distribution $\mathbb{P}(X_t=x)$ as $t\to\infty$ in the next section, concentrating on values of time that are of the form $3t$. Other values of time show an undistinguishable behavior (see also Appendix~\ref{app:3t+1_and_3t+2}).
\begin{figure}[h]
\begin{center}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.7]{fig1-a.eps}\\[2mm]
(a) $\theta=\frac{\pi}{4}$
\end{center}
\end{minipage}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.7]{fig1-b.eps}\\[2mm]
(b) $\theta=\frac{2\pi}{5}$
\end{center}
\end{minipage}
\vspace{5mm}
\fcaption{Time evolution of probability distributions in the case of $\alpha=1/\sqrt{2}, \beta=i/\sqrt{2}$}
\label{fig:time-probability}
\end{center}
\end{figure}
\begin{figure}[h]
\begin{center}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.7]{fig2-a.eps}\\[2mm]
(a) $\alpha=1/\sqrt{2},\,\beta=i/\sqrt{2}$
\end{center}
\end{minipage}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.7]{fig2-b.eps}\\[2mm]
(b) $\alpha=1,\,\beta=0$
\end{center}
\end{minipage}
\vspace{5mm}
\fcaption{The relationships between the probability distribution at time $t=150$ and the parameter $\theta$ which determines the coin-flip operator $C$}
\label{fig:theta-probability}
\end{center}
\end{figure}
\section{Long-time limit theorem and its proof}
\label{sec:limit_th}
We get a long-time limit theorem for the probability distribution and its proof in this section assuming that the walker starts from the origin.
Let us take an initial state $\ket{\Psi_0}=\ket{0}\otimes\left(\alpha\ket{0}+\beta\ket{1}\right)$ with $|\alpha|^2+|\beta|^2=1$.
This initial condition means that the walker starts from the origin because of $\mathbb{P}(X_0=0)=1$.
Then we obtain a limit theorem for the 3-period time-dependent QW.
\begin{thm}
\begin{align}
\lim_{t\to\infty}\mathbb{P}\left(\frac{X_{3t}}{3t}\leq x\right)
=\int_{-\infty}^x \Biggl[&\left\{1-\nu(\alpha,\beta; y)\right\}f(y)I_{\left(\frac{1-4c^2}{3},\frac{\sqrt{1+8c^2}}{3}\right)}(y) \nonumber\\
&+\left\{1+\nu(\alpha,\beta; -y)\right\}f(-y)I_{\left(-\frac{\sqrt{1+8c^2}}{3},-\frac{1-4c^2}{3}\right)}(y)\Biggr]\,dy,\label{eq:cumulative}
\end{align}
where
\begin{align}
f(x)=&\frac{|s|\left(|s|x+\sqrt{D(x)}\right)^2}{\pi(1-x^2)\sqrt{W_{+}(x)}\sqrt{W_{-}(x)}\sqrt{D(x)}},\\[3mm]
\nu(\alpha,\beta; x)=&\frac{1}{c(1+8c^2)}\left\{9c^3(|\alpha|^2-|\beta|^2)+3s(1+6c^2)\Re(\alpha\overline{\beta})\right\}x\nonumber\\
&+\frac{s}{c|s|(1+8c^2)}\left\{cs(|\alpha|^2-|\beta|^2)-(1+2c^2)\Re(\alpha\overline{\beta})\right\}\sqrt{D(x)},\\[3mm]
D(x)=&1+8c^2-9c^2x^2,\\
W_{+}(x)=&-(1-4c^2)+3(1-2c^2)x^2+2|s|x\sqrt{D(x)},\\
W_{-}(x)=&1+8c^2-3(1+2c^2)x^2-2|s|x\sqrt{D(x)},\\[2mm]
I_A(x)=&\left\{\begin{array}{cl}
1&(x\in A)\\
0&(x\notin A)
\end{array}\right.,
\end{align}
and $\Re(z)$ denotes the real part of the complex number $z$.
\label{th:limit}
\end{thm}
The function $\nu(\alpha,\beta; x)$ is the part of the limit density function which gives the effect of the initial condition $\alpha,\beta$ on the limit behavior, and if the conditions $|\alpha|=|\beta|$ and $\Re(\alpha\overline{\beta})=0 $ are satisfied simultaneously (e.g. $\alpha=1/\sqrt{2}, \beta=i/\sqrt{2}$\,), this term disappears.
Note that $D(x), W_{+}(x), W_{-}(x) >0$ for $x\in \left(-\frac{\sqrt{1+8c^2}}{3},-\frac{1-4c^2}{3}\right) \cup \left(\frac{1-4c^2}{3},\frac{\sqrt{1+8c^2}}{3}\right)$ and $|1-4c^2|<\sqrt{1+8c^2}$.
As examples, Fig.~\ref{fig:limit} shows probability distributions and the limit density functions when $\alpha=1/\sqrt{2},\, \beta=i/\sqrt{2}$.
\begin{figure}[h]
\begin{center}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.5]{fig3-a.eps}\\[2mm]
(a) $\theta=\frac{\pi}{4}$
\end{center}
\end{minipage}
\begin{minipage}{70mm}
\begin{center}
\includegraphics[scale=0.5]{fig3-b.eps}\\[2mm]
(b) $\theta=\frac{2\pi}{5}$
\end{center}
\end{minipage}
\vspace{5mm}
\fcaption{Probability distribution at time $999\,(=3\times 333)$ (blue line) and the limit density function (red line), in the case of $\alpha=1/\sqrt{2}, \beta=i/\sqrt{2}$}
\label{fig:limit}
\end{center}
\end{figure}
\vspace{5mm}
\begin{proof}{
To prove the limit theorem we use Fourier analysis in the way introduced in Grimmett et al.~\cite{GrimmettJansonScudo2004}, and derive a convergence of the $r$-th moment $\mathbb{E}\left[(X_{3t}/3t)^r\right]$ ($r=0,1,2,\ldots$) which is equivalent to a convergence of the generating function $\mathbb{E}[e^{izX_{3t}/3t}]$.
First, we consider the following Fourier transform $\ket{\hat\Psi_t(k)}\, (k\in [-\pi,\pi))$ derived from the states of the walker
\begin{equation}
\ket{\hat\Psi_t(k)}=\sum_{x\in\mathbb{Z}}e^{-ikx}\ket{\psi_t(x)}.
\end{equation}
We should note that we can obtain the state $\ket{\psi_t(x)}$ by using the inverse Fourier transform
\begin{equation}
\ket{\psi_t(x)}=\int_{-\pi}^\pi e^{ikx}\ket{\hat\Psi_t(k)}\frac{dk}{2\pi}.
\end{equation}
Equation~(\ref{eq:time-evolution}) produces a time evolution of the Fourier transform
\begin{align}
\ket{\hat\Psi_{3t}(k)}=&\left(\hat S(k)\hat C(k)^2\right)^t\ket{\hat\Psi_0(k)},\nonumber\\
\ket{\hat\Psi_{3t+1}(k)}=&\hat C(k)\left(\hat S(k)\hat C(k)^2\right)^t\ket{\hat\Psi_0(k)},\\
\ket{\hat\Psi_{3t+2}(k)}=&\hat C(k)^2\left(\hat S(k)\hat C(k)^2\right)^t\ket{\hat\Psi_0(k)},\nonumber
\end{align}
where $\hat S(k)=e^{ik}\ket{0}\bra{0}+e^{-ik}\ket{1}\bra{1}$ and $\hat C(k)=\hat S(k)C$.
The operator $\hat S(k)$ corresponds to the position-shift operator $\tilde{S}$.
Before computing the $r$-th moment $\mathbb{E}\left(X_{3t}^r\right)$, we get the eigenvalues and the normalized eigenvectors of the unitary matrix $\hat S(k)\hat C(k)^2$ so that we rewrite the Fourier transform $\ket{\hat\Psi_{3t}(k)}$ on the appropriate eigenspace.
Let us take a standard basis as the orthogonal normalized basis $\left\{\ket{0},\ket{1}\right\}$ with
\begin{equation}
\ket{0}=\left[\begin{array}{c}
1\\0
\end{array}\right],\quad
\ket{1}=\left[\begin{array}{c}
0\\1
\end{array}\right].
\end{equation}
Then the matrix $\hat S(k)\hat C(k)^2$ has two eigenvalues
\begin{equation}
\lambda_j(k)=c^2\cos 3k+s^2\cos k -(-1)^j i\sqrt{1-(c^2\cos 3k+s^2\cos k)^2} \quad (j=1,2),
\end{equation}
and they are distinct as long as $k\neq -\pi,0$.
Again, we should note that $1-(c^2\cos 3k+s^2\cos k)^2$ is not a negative number and its value is zero if and only if $k= -\pi,0$.
As one of the possible expressions of the normalized eigenvector corresponding to each eigenvalue $\lambda_j(k)$, we have
\begin{equation}
\ket{v_j(k)}=\frac{1}{\sqrt{N_j(k)}}\left[\begin{array}{c}
-2cs\,e^{2ik}\sin k\\[2mm]
c^2\sin 3k+s^2\sin k+(-1)^j\sqrt{1-(c^2\cos 3k+s^2\cos k)^2}
\end{array}\right],
\end{equation}
where $N_{j}(k)$ are normalization factors given by
\begin{align}
N_{j}(k)=&2\biggl\{1-(c^2\cos 3k+s^2\cos k)^2 \nonumber\\
&+(-1)^j(c^2\sin 3k+s^2\sin k)\sqrt{1-(c^2\cos 3k+s^2\cos k)^2}\biggr\}.
\end{align}
Here, we treat the $r$-th moments at time $3t$ and express them in the Fourier space by using the eigenvalues $\lambda_j(k)$ and the eigenvectors $\ket{v_j(k)}$.
With a decomposition $\ket{\hat\Psi_{3t}(k)}=\sum_{j=0}^1\lambda_j^t(k)\braket{v_j(k)|\hat\Psi_0(k)}\ket{v_j(k)}$, we get
\begin{align}
\mathbb{E}(X_{3t}^r)=&\sum_{x\in\mathbb{Z}}x^r\mathbb{P}(X_{3t}=x)\nonumber\\
=&\int_{-\pi}^\pi \bra{\hat\Psi_{3t}(k)}\left(D^r\ket{\hat\Psi_{3t}(k)}\right)\frac{dk}{2\pi}\nonumber\\
=&(t)_r\int_{-\pi}^\pi \sum_{j=1}^2 \left(\frac{i\lambda'_j(k)}{\lambda_j(k)}\right)^r\left|\braket{v_j(k)|\hat\Psi_0(k)}\right|^2\frac{dk}{2\pi}+O(t^{r-1}),
\label{eq:r-th_moment}
\end{align}
where $D=i(d/dk)$ and $(t)_r=t(t-1)\times\cdots\times(t-r+1)$.
Equation~(\ref{eq:r-th_moment}) gives us a convergence as $t\to\infty$,
\begin{equation}
\lim_{t\to\infty}\mathbb{E}\left[\left(\frac{X_{3t}}{3t}\right)^r\right]=\int_{-\pi}^\pi \sum_{j=1}^2 \left(\frac{i\lambda'_j(k)}{3\lambda_j(k)}\right)^r\left|\braket{v_j(k)|\hat\Psi_0(k)}\right|^2\frac{dk}{2\pi},
\label{eq:r-th_moment2}
\end{equation}
where
\begin{equation}
\frac{i\lambda'_j(k)}{3\lambda_j(k)}=(-1)^j \frac{3c^2\sin 3k+s^2\sin k}{3\sqrt{1-(c^2\cos 3k+s^2\cos k)^2}}.
\end{equation}
Setting $i\lambda'_j(k)/3\lambda_j(k)=x$ in Eq.~(\ref{eq:r-th_moment2}) takes us to our goal because we have
\begin{align}
\lim_{t\to\infty}\mathbb{E}\left[\left(\frac{X_{3t}}{3t}\right)^r\right]=&\int_{-\infty}^\infty x^r \Biggl[\left\{1-\nu(\alpha,\beta; x)\right\}f(x)I_{\left(\frac{1-4c^2}{3},\frac{\sqrt{1+8c^2}}{3}\right)}(x) \nonumber\\
&+\left\{1+\nu(\alpha,\beta; -x)\right\}f(-x)I_{\left(-\frac{\sqrt{1+8c^2}}{3},-\frac{1-4c^2}{3}\right)}(x)\Biggr]\,dx,
\end{align}
which means that the random variable $X_{3t}/3t$ converges in distribution to a random variable with a density function
\begin{equation}
\left\{1-\nu(\alpha,\beta; x)\right\}f(x)I_{\left(\frac{1-4c^2}{3},\frac{\sqrt{1+8c^2}}{3}\right)}(x)+\left\{1+\nu(\alpha,\beta; -x)\right\}f(-x)I_{\left(-\frac{\sqrt{1+8c^2}}{3},-\frac{1-4c^2}{3}\right)}(x).
\end{equation}
To obtain the cumulative distribution function on the left hand side of Eq.~(\ref{eq:cumulative}), we need to integrate this density.
}
\end{proof}
\section{Summary and Discussion}
\label{sec:summary}
We have dealt with a 3-period time-dependent discrete-time 2-state QW on the line with the walker located at the origin at the initial time, and gave a limit theorem which gives the asymptotic behavior of the walker after a large number of steps.
On a rescaled space by time, the position of the walker converges in distribution to a random variable. The density function of the random variable has a compact support.
Its shape resembles that of a doubled arcsine distribution.
When we choose the parameter $\theta$, which determines the coin-flip operator $C$, in the open interval $(\pi/3, 2\pi/3) \cup (4\pi/3, 5\pi/3)$, we do not observe the walker at the starting point after a long time as shown in Fig.~\ref{fig:limit}-(b) because the compact support is the open interval $\left(-\frac{\sqrt{1+8\cos^2\theta}}{3},-\frac{1-4\cos^2\theta}{3}\right)\cup\left(\frac{1-4\cos^2\theta}{3},\frac{\sqrt{1+8\cos^2\theta}}{3}\right)$.
For a time-independent walk or a 2-period time-dependent walk starting from the origin, the initial condition at the origin produces a linear function in their limit density functions~\cite{Konno2002a,MachidaKonno2010}.
On the other hand, the 3-period time-dependent walk treated in this paper, features a non-linear term reflecting the initial condition at the origin, which is expressed by $\nu(\alpha,\beta; x)$ in the limit theorem, and the function $\sqrt{D(x)}=\sqrt{1+8c^2-9c^2x^2}$.
We showed that the limit distribution of the 3-period time-dependent walk is essentially different from that of the time-independent walk or the 2-period time-dependent walk.
We have treated a 3-period time-dependent walk whose coin-state is flipped by only one coin-flip operator $C$ at time
$t=0,1 \mod 3$, and is shifted without any coin-flip operation at time $t=2 \mod 3$.
We can also see very interesting behavior for a 3-period time-dependent walk with three distinct coin-flip operators by using numerics as displayed in the appendix. We intend to analyze these results carefully in a future publication.
We have described a mathematical property of a 3-period time-dependent walk.
It would be worth discussing this phenomenon from the perspective of physics, for example it would be nice to explore a possible application to the design of selective pulses in~\cite{MorrisMcIntyreRourkeNgo1989}.
\nonumsection{Acknowledgements}
\noindent T. Machida is grateful to the Japan Society for the Promotion of Science for the support, and to the Math. Dept. UC Berkeley for hospitality.
F.A. Gr\"{u}nbaum acknowledges support from the Applied Math. Sciences subprogram of the Office of Energy Research, US Department of Energy, under
Contract DE-AC03-76SF00098, and from AFOSR grant FA95501210087 through a subcontract to Carnegie Mellon University.
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 304
|
Q: what are the caveats of sending email via the "mail" command? I want to send emails from my work computers so that I can notify myself when various long-running tasks are completed. As I understand it, the command below looks up bar.com's MX record, makes a connection to the mail server and does SMTP to send the message:
echo "Hello world" | mail -s "Hello world" foo@bar.com
What do I need to know so that I can avoid being flagged/blocked as a spammer?
A: There is no difference between sending mail via the mail command or via any other program. As such a mail send via mail (1) is not more nor less likely to be identified as spam.
( I would add that this is the default way in which many non-cron tasks send you their mail, but I have no evidence to back that up. )
As to avoid having your mail seen as spam: Make sure that your mail does not look like spam.
E.g. not just a single HTML link, not just a picture. No l33t spelling. Valid origins. Etc etc. Non of these are specific to the mail command.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,696
|
package org.gedcomx.util.date;
import junit.framework.TestCase;
/**
* Class for testing the GedcomX FormalDate utility class.
* User: Randy Wilson
* Date: 8/20/13
* Time: 2:28 PM
*/
public class FormalDateTest extends TestCase {
/*
* Simple date:
* (+|-)YYYY[-MM[-DD[Thh[:mm[:ss[(+\-)hh[:mm]|Z]]]]]]
*
* Duration:
* P[yyyyY][mmM][ddD][T[hhH][mmM][ssS]]
*
* Closed date Range:
* [simpleDate]/[simpleDate|Duration]
* Open-ended date range:
* [simpleDate]/
* /[simpleDate]
* Recurring date
* R[repetitions]/simpleDate/(simpleDate|Duration)
* Approximate date or date range
* A(simpleDate)
* A(dateRange)
*
* => ([A](simpleDate|dateRange) | R[repetitions]/simpleDate/(simpleDate|Duration))
* => [A]simpleDate
* [A]simpleDate/[simpleDate|Duration]
* [A]/simpleDate
* R[repetitions]/simpleDate/(simpleDate|Duration)
*/
public void testFormalDate() {
// Simple date
checkFormalDate("+1820", false, 1820, false, null, null, null);
checkFormalDate("A+1820", true, 1820, false, null, null, null);
failFormalDate("1820"); // no "+"
failFormalDate("+1820--12"); // no month
// Ranges
checkFormalDate("A+1820/", true, 1820, true, null, null, null);
checkFormalDate("A+1820/+1830", true, 1820, true, 1830, null, null);
checkFormalDate("A/+1830", true, null, true, 1830, null, null);
checkFormalDate("/+1830", false, null, true, 1830, null, null);
// Duration
checkFormalDate("+1820/P0061Y", false, 1820, true, null, 61, null);
checkFormalDate("A+1820/P0061Y", true, 1820, true, null, 61, null);
failFormalDate("/P0061Y"); // no start date
failFormalDate("A+1820/P61Y"); // need to 0-pad duration
// Repetition
checkFormalDate("R/+1820/+1821", false, 1820, true, 1821, null, 0);
checkFormalDate("R42/+1820/+1821", false, 1820, true, 1821, null, 42);
checkFormalDate("R42/+1820/P0001Y", false, 1820, true, null, 1, 42);
checkFormalDate("R/+1820/P0061Y", false, 1820, true, null, 61, 0);
failFormalDate("AR/+1820/P0061Y"); // can't have approximate recurring date
failFormalDate("RA/+1820/P0061Y"); // can't have approximate recurring date
}
private void failFormalDate(String formalDateString) {
try {
new FormalDate(formalDateString);
fail("Should have failed to parse formal date string: " + formalDateString);
}
catch (IllegalArgumentException e) {
// ok
}
}
private FormalDate checkFormalDate(String formalDateString, boolean isApproximate, Integer startYear, boolean isRange, Integer endYear, Integer durationYears, Integer repetitions) {
FormalDate formalDate = new FormalDate(formalDateString);
assertEquals(isApproximate, formalDate.isApproximate());
if (startYear == null) {
assertNull(formalDate.getStart());
}
else {
checkSimpleDate(formalDate.getStart(), startYear);
}
assertEquals(isRange, formalDate.isRange());
if (endYear == null) {
assertNull(formalDate.getEnd());
}
else {
checkSimpleDate(formalDate.getEnd(), endYear);
}
if (durationYears == null) {
assertNull(formalDate.getDuration());
}
else {
assertEquals(durationYears, formalDate.getDuration().getYear());
}
if (repetitions == null) {
assertFalse(formalDate.isRecurring());
}
else {
assertTrue(formalDate.isRecurring());
if (repetitions == 0) {
assertNull(formalDate.getNumRepetitions());
}
else {
assertEquals(repetitions, formalDate.getNumRepetitions());
}
}
assertEquals(formalDateString, formalDate.toString());
assertTrue(formalDate.isValid());
return formalDate;
}
public void testSimpleDate() {
FormalDate formalDate = roundTrip("+1830-04-06");
checkSimpleDate(formalDate.getStart(), 1830, 4, 6, null, null, null, false, null, null);
formalDate = new FormalDate("A+1820-12-31T23:59:01Z");
assertTrue(formalDate.isApproximate());
checkSimpleDate(formalDate.getStart(), 1820,12, 31, 23, 59, 1, true, null, null);
}
private void checkSimpleDate(SimpleDate simpleDate, int year) {
checkSimpleDate(simpleDate, year, null, null, null, null, null, false, null, null);
}
private void checkSimpleDate(SimpleDate simpleDate, int year, Integer month, Integer day, Integer hour, Integer minute, Integer second, boolean isUTC, Integer tzHour, Integer tzMinute) {
checkInt(year, simpleDate.getYear());
checkInt(month, simpleDate.getMonth());
checkInt(day, simpleDate.getDay());
checkInt(hour, simpleDate.getHour());
checkInt(minute, simpleDate.getMinute());
checkInt(second, simpleDate.getSecond());
checkInt(tzHour, simpleDate.getTimeZoneHours());
checkInt(tzMinute, simpleDate.getTimeZoneMinutes());
assertEquals(isUTC, simpleDate.isUTC());
if (isUTC) {
assertNull(simpleDate.getTimeZoneHours());
assertNull(simpleDate.getTimeZoneMinutes());
}
}
private void checkInt(Integer a, Integer b) {
if (a == null) {
assertNull(b);
}
else {
assertNotNull(b);
assertEquals(a.intValue(), b.intValue());
}
}
private FormalDate roundTrip(String formalDateString) {
FormalDate formalDate = new FormalDate(formalDateString);
assertEquals(formalDateString, formalDate.toString());
return formalDate;
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,793
|
The GOP has taken a number of swings at cutting benefits for families who rely on Supplemental Nutrition Assistance Programs, or SNAP, to help meet their food needs. State legislatures have demanded that work requirements be met to continue receiving often scant benefits. Paul Ryan suggested converting the funds to block grants for the states to dole out. And one legislator even demanded that fish not be allowed for purchase.
Now the Republican Party is so enamored with the myth that those who use EBT cards are living large on taxpayer dollars that even President Donald Trump is in on the game. In the most recent budget proposal for 2019, President Trump recommends cutting the SNAP budget by $17.2 billion just for 2019 — over 20 percent of the program's cost.
Except, there's an obvious loss in food benefits: the ability and autonomy to make one's own food choices.
Currently, SNAP recipients average around $125 a month for food purchases, or $254 for a household — by no means a large amount from the start. The new budget proposal would reduce the actual discretionary amount to about $63/$125 dollars for an entire month, or approximately $2/4 a day.
For those already finding it nearly impossible to stretch their monthly limit, this reduction in funds will be even more problematic. Meanwhile, the government plans to provide pre-filled boxes with cheap and far less healthy food. Cereals are high in sugar, canned fruits and vegetables have far less nutrients then fresh or frozen and canned beans and meats are loaded with additional salts to preserve them.
The closest analogy that the federal government runs now is a program that distributes food boxes instead of SNAP benefits on American Indian reservations. Available foods include canned beans, corn, and spinach; dried eggs; dried and canned fruit; peanut butter; and canned and frozen meat. It's a far cry from a meal kit box with fresh grass-fed beef, kale, and a tiny container of Sriracha. (And also serves far fewer people — about 90,000.).
Despite claims otherwise, the SNAP program is almost entirely without "fraud" and the options for obtaining "unhealthy" food have already been dramatically limited by restrictions placed on the funds. The Trump administration's plan wouldn't ensure that more healthy food is purchased; it would remove a person's ability buy fresh foods like dairy, produce and fresh and frozen meat — especially in areas of the country where local farmer's markets allow EBT cards for purchases of local fresh goods in season and at cheaper costs than in stores.
Those who use SNAP don't need pre-packaged boxes or reduced benefits. What they need is the autonomy to purchase the food that best fits their family's needs and tastes — and our trust that they know what is best for their household.
If the government truly wants beneficiaries to eat healthier foods, they need to give people more money for better food.
i AGREE. AMERICA IS RICH ENOUGH TO HELP OUR POOR ALOT MORE!!
The food boxes would undoubtedly contain unhealthy food, most likely things near the expiration date. Trust people to make their own judgments, and by all means, raise the amount of money on the food stamp cards. One thing we can definitely guarantee, is that lawmakers proposing this change don't eat the cheap food they want to give to the poor.
So lets go to place & get MORE food boxes of food for FREE to help us out. & black women & girls please DON'T & stop having abortions & taking birth control as that IS black genocide & they are making the birth control from the horses urine Exterminating them as well. Please listen to part 13 1st of the film Maafa21 in which Dr Martin Luther King jr niece speaks. & please have 2 babies with a black MAN black women as it takes 2 people to replace another 2 people,& interracial mixing & Marriage is the number ONE thing that IS Exterminating us black people WORLDWIDE even more than Abortion. Jesus was Black & Black IS Beautiful so please CHANGE & LOVE yourselves black people.
Yeah us people who need & get food stamps should get more food stamps & the minimum wage that I voted on to be increased should be increased MORE as the cost of living is so high.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 8,260
|
Designed by MUTE. Manufactured by HATASHIKKITE, the col. series are products of creative Japanese inspiration made using the traditional craft technique of 'Rokuro Biki' (wood turning).
Available in beautiful colour combinations, these Stacking Containers are useful for storage and platters for food when entertaining. Each piece is unique, with the wood characteristics becoming an integral part of the container itself.
While these Stacking Containers look gorgeous on your shelf in the kitchen, and can be used to serve and carry food, the containers can also be used as storage for jewellery, bathroom cotton balls and other essentials, bedside objects and a myriad of other bibs and bobs!
Hatashikkiten was established in 1930 by Hata Unomatsu, who was himself a craftsman of Yamanaka Shikki (lacquer ware). Commitment to quality, selection of materials without compromise and masterful techniques have all been passed on until today.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,921
|
{"url":"https:\/\/reason.town\/f1-loss-pytorch\/","text":"# F1 Loss in Pytorch\n\nF1 Loss in Pytorch \u2013 This is a blog post about the F1 Loss function in Pytorch.\n\nCheckout this video:\n\n## Introduction\n\nIn this tutorial, we\u2019ll learn about the F1 loss function in PyTorch. The F1 loss function is commonly used in classification tasks. It\u2019s a combination of the precision and recall scores. The precision score is the number of True Positives divided by the sum of True Positives and False Positives. The recall score is the number of True Positives divided by the sum of True Positives and False Negatives. The F1 score is the harmonic mean of the precision and recall scores.\n\n## What is Pytorch?\n\nPytorch is a Python-based scientific computing package that uses the power of graphics processing units. It is frequently used by researchers in the field of deep learning and computer vision.\n\n## What is an F1 loss?\n\nF1 loss is a type of loss function that is used in classification tasks. It is typically used when there are two classes, but can be extended to multiple classes. The function calculates the mean of the precision and recall for each class and then takes the harmonic mean of those values. This gives rise to the name \u201cF1\u201d, which stands for \u201charmonic mean of the precision and recall\u201d.\n\nThe F1 loss is usually used in conjunction with a softmax layer in order to calculate the cross-entropy loss for a multi-class classification task. The cross-entropy loss is then minimized in order to find the model parameters that result in the best classification performance.\n\n## How can Pytorch be used to calculate an F1 loss?\n\nWhile there is no built-in F1 loss function in Pytorch, it is possible to calculate an F1 loss using the standard BCEWithLogitsLoss and ignore_index arguments. Using BCEWithLogitsLoss as the base loss function, we can first calculate the cross entropy over all classes except for the one being predicted. We can then add a term to penalize incorrect predictions of that class. The ignore_index argument can be used to tell BCEWithLogitsLoss to ignore the cross entropy calculation for the class being predicted, while still performing backpropagation through the rest of the network.\n\n## What are some benefits of using Pytorch to calculate an F1 loss?\n\nPytorch is a powerful tool for deep learning that can be used to calculate an F1 loss. There are many benefits to using Pytorch to calculate an F1 loss, including the ability to parallelize code, the ability to use GPUs for training, and the ability to easily change network architectures.\n\n## What are some potential drawbacks of using Pytorch to calculate an F1 loss?\n\nSome potential drawbacks of using Pytorch to calculate an F1 loss are as follows:\n\n1. Pytorch is relatively new, and as such, there is not a lot of documentation or support available for it.\n\n2. Pytorch is not as widely used as some of the other options out there, so there may be less people familiar with it and therefore less able to help with any issues that may arise.\n\n3. Because Pytorch is still new, it is constantly evolving and changing, which can make it difficult to keep up with the latest changes.\n\n## How can Pytorch be used to improve the accuracy of an F1 loss calculation?\n\nWhile there are many ways to improve the accuracy of an F1 loss calculation, one common way is to use Pytorch. Pytorch is a machine learning library that provides tools for efficient calculations and can be used to improve the accuracy of predictions made by a model. When used in conjunction with other methods, Pytorch can help to improve the accuracy of an F1 loss calculation by up to 30%.\n\n## What are some other potential applications for Pytorch?\n\nThere are many potential applications for Pytorch. Some of these include:\n\n-F1 metric for ranking and classification problems\n-Training and deploying deep learning models on GPUs\n-Optimizing hyperparameters for deep learning models\n\u2013 debugging Neural Networks\n\n## Conclusion\n\nF1-score is a popular metric for evaluating the performance of classification models. It is a combination of precision and recall. Precision is the ratio of correctly predicted positive observations to the total predicted positive observations. Recall is the ratio of correctly predicted positive observations to the all actual positive observations in the data.\n\nF1-score = 2*(Recall * Precision) \/ (Recall + Precision)\n\nIf we have a balanced dataset, meaning equal numbers of observations for each class, then accuracy will be an appropriate metric to use. However, in many real-world datasets, this is not often the case. In these situations, using precision and recall together can give us a better idea of how our model is performing.\n\nIn Pytorch, there is no built-in F1-score metric, so we will need to create one. We can do this by subclassing Pytorch\u2019s nn.Module class and creating our own F1Loss class.\n\nOur F1Loss class will take in two inputs: predictions and targets. The predictions should be probabilities for each class (output from our model), and the targets should be the true labels for each observation. We will then calculate the precision and recall for each class and take the weighted average to get our final F1-score loss value.\n\nclass F1Loss(nn.Module):\ndef __init__(self):\nsuper().__init__()\n\ndef forward(self, predictions, targets):\n\n# calculate precision and recall for each class\n\n# take weighted average to get final loss value\n\nreturn loss\n\n## References\n\n\u2013 https:\/\/blog.floydhub.com\/a-beginners-guide-to-loss-functions-in-machine-learning\/\n\u2013 https:\/\/towardsdatascience.com\/commonly-used-loss-functions-in-machine learning7e0ed9f23ce1\n\nIn machine learning, we frequently encounter different types of data and problems. To address these different types of data and problems, there exist various types of loss functions. In this blog post, we will briefly review some of the most commonly used loss functions in machine learning.\n\nThe first loss function that we will discuss is the mean squared error (MSE) loss. MSE is commonly used when we are dealing with regression problems. The MSE loss is defined as:\n\n$$\\mathrm{MSE}(\\hat{y}, y) = \\frac{1}{n}\\sum_{i=1}^{n}(\\hat{y}_i \u2013 y_i)^2$$\n\nwhere $\\hat{y}$ is the predicted value and $y$ is the true value. The MSE loss punishes predicted values that are far from the true values more than those that are close to the true values. This makes sense since we usually care more about getting the predicted values close to the true values.\n\nAnother popular loss function is the cross entropy loss, which is frequently used when dealing with classification problems. The cross entropy loss is defined as:\n\n$$\\mathrm{CE}(\\hat{y}, y) = -\\frac{1}{n}\\sum_{i=1}^{n}\\left(y_i \\log \\hat{y}_i + (1 \u2013 y_i)\\log (1 \u2013 \\hat{y}_i)\\right)$$\n\nwhere $\\hat{y}$ is the predicted probability of the class being 1 and $y$ is the true label (0 or 1). The cross entropy loss punishes predicted probabilities that are far from the true labels more than those that are close to the true labels. This again makes sense since we usually care more about getting the predicted probabilities close to the true labels.\n\nThere are many other types of losses out there, but these are two of the most commonly used ones. In general, it is important to carefully select a loss function that makes sense for your specific problem and data set.\n\nScroll to Top","date":"2022-11-30 16:40:28","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 1, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.5847043991088867, \"perplexity\": 446.50285092123306}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 5, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2022-49\/segments\/1669446710765.76\/warc\/CC-MAIN-20221130160457-20221130190457-00823.warc.gz\"}"}
| null | null |
A Wanderer in Venice
by E.V. Lucas
Previous Part 1 2 3 4 5 6 7 Next Part
Home - Random Browse
With the best will to do so, I cannot be much impressed by the glory and power of the Doges. They wear a look, to me, very little removed from Town Councillors: carried out to the highest power, no doubt, but incorrigibly municipal none the less; and the journey through these halls of their deliberations is tedious and unenchanting. That I am wrong I am only too well aware. Does not Venetian history, with its triumphs and pageantry of world-power, prove it? And would Titian and Paul Veronese and Tintoretto have done all this for a Mayor and Corporation? These are awkward questions. None the less, there it is, and the Doges' Palace, within, would impart no thrill to me were it not for Tintoretto's "Bacchus and Ariadne."
Having paid for our tickets (for only on Sundays and holidays is the Palace free) we take the Scala d'Oro, designed by Sansovino, originally intended only for the feet of the grandees of the Golden Book. The first room is an ante-room where catalogues are sold; but these are not needed, for every room, or nearly every room, has hand-charts of the paintings, and every room has a custodian eager to impart information. Next is the Hall of the Four Doors, with its famous and typical Titian—Doge Grimani, fully armed and accompanied by warriors, ecstatically acknowledging religion, as symbolized by a woman, a cross, and countless cherubim. Behind her is S. Mark with an expression of some sternness, and beside him his lion, roaring.
Doges, it appears,—at any rate the Doges who reigned during Titian's long life—had no sense of humour, or they could not have permitted this kind of self-glorification in paint. Both here and at the Accademia we shall see picture after picture in which these purse-proud Venetian administrators, suspecting no incongruity or absurdity, are placed, by Titian and Tintoretto, on terms of perfect intimacy with the hierarchy of heaven. Sometimes they merely fraternize; sometimes they masquerade as the Three Kings or Wise Men from the East; but always it is into the New Testament that, with the aid of the brush of genius, they force their way.
Modesty can never have been a Venetian characteristic; nor is it now, when Venice is only a museum and show place. All the Venetians—the men, that is,—whom one sees in the Piazza have an air of profound self-satisfaction. And this palace of the Doges is no training-place for humility; for if its walls do not bear witness, glorious and chromatic, to the greatness of a Doge, it is merely because the greatness of the Republic requires the space. In this room, for example, we find Tiepolo allegorizing Venice as the conqueror of the sea.
And now for the jewel of art in the Doges' Palace. It is in the room opposite the door by which we entered—the ante-room of the Sala del Collegio—and it faces us, on the left as we enter: the "Bacchus and Ariadne" of Tintoretto. We have all seen the "Bacchus and Ariadne" of Titian in our National Gallery, that superb, burning, synchronized epitome of the whole legend. Tintoretto has chosen one incident only; Love bringing Bacchus to the arms of Ariadne and at the same moment placing on his head a starry coronal. Even here the eternal pride of Venice comes in, for, made local, it has been construed as Love, or say Destiny, completing the nuptials of the Adriatic (Bacchus) with Venice (Ariadne), and conferring on Venice the crown of supremacy. But that matters nothing. What matters is that the picture is at once Tintoretto's simplest work and his most lovely. One can do nothing but enjoy it in a kind of stupor of satisfaction, so soothing and perfect is it. His "Crucifixion," which we shall see at the Scuola of S. Rocco, must ever be this giant painter's most tremendous achievement; but the picture before us must equally remain his culminating effort in serene, absolute beauty. Three other mythological paintings, companions of the "Bacchus," are here too, of which I like best the "Minerva" and the "Mercury"; but they are far from having the quality of that other. I have an idea that "The Origin of the Milky Way," in the National Gallery, was painted as a ceiling piece to go with these four, but I have no data for the theory, beyond its similarity in size and scheme. The other great picture in this room is Paul Veronese's sumptuous "Rape of Europa."
The Sala del Collegio itself, leading from this room, is full of Doges in all the magnificence of paint, above the tawdriest of wainscotting. Tintoretto gives us Doge Andrea Gritti praying to the Virgin, Doge Francesco Donato witnessing as an honoured guest the nuptials of S. Catherine, Doge Niccolo da Ponte surveying the Virgin in glory, and Doge Alvise Mocenigo condescending to adore his Saviour. Paul Veronese depicts an allegory of the battle of Lepanto in 1571, at which Venice temporarily overcame the Turks. The kneeling white-bearded warrior beside S. Giustina is the victor, afterwards Doge Sebastiano Venier, and Christ looks on in approval. Tintoretto also painted for the Palace a picture of this battle, but it perished in the fire of 1576. It is Veronese who painted the virtues and attributes on the ceiling, one of his most famous works being the woman with a web, who is sometimes called "Industry" and sometimes "Dialectics," so flexible is symbolism. "Fidelity" has a dog with a fine trustful head. To my weary eye the finest of the groups is that of Mars and Neptune, with flying cherubs, which is superbly drawn and coloured. Nothing but a chaise-longue on which to lie supine, at ease, can make the study of these wonderful ceilings anything but a distressing source of fatigue.
The next room is the Sala del Senato, and here again we find a blend of heaven and Venice, with Doges as a common denominator. A "Descent from the Cross" (by Tintoretto) is witnessed by Doge Pietro Lando and Doge Marcantonio Trevisan; and the same hand gives us Pietro Loredan imploring the aid of the Virgin. In the centre ceiling painting Tintoretto depicts Venice as Queen of the Sea. The other artist here is Palma the younger, whose principal picture represents Doge Leonardo Loredan presiding over an attack by a lion on a bull, typifying the position of the Republic when Pope Julius launched the League of Cambray against it in 1508. The Doge does not look dismayed, but Venice never recovered from the blow.
The room on the right of the throne leads to the chapel, which has several small pictures. A Giovanni Bellini is over the altar, but it is not one of his best. During his long life in Venice Bellini saw ten Doges, and in his capacity as ducal painter painted four of them.
Returning to the Sala delle Quattro Porte (by way of the "Bacchus and Ariadne" room, if we are wise), we make for the Sala del Consiglio dei Dieci, the terrible Council of Ten. All Venetian histories are eloquent upon this secret Tribunal, which, more powerful far than the Doge himself, for five centuries, beginning early in the fourteenth, ruled the city. On the walls are historical paintings which are admirable examples of story-telling, and on the ceiling are Veroneses, original or copied, the best of which depicts an old man with his head on his hand, fine both in drawing and colour. It was in the wall of the next room that the famous Bocca di Leone was placed, into which were dropped those anonymous charges against Venetian citizens which the Council of Ten investigated, and if true, or, very likely, if not true, punished with such swiftness and thoroughness. How a state that offered such easy temptations to anti-social baseness and treachery could expect to prosper one cannot imagine. It suggests that the Venetian knowledge of human nature was defective at the roots.
In the next room the Three Heads of the Council of Ten debated, and here the attendant goes into spasms of delight over a dazzling inlaid floor.
This is all that is shown upstairs, for the piombi, or prison cells in the leaden roof, are now closed.
Downstairs we come to the two Great Halls—first the gigantic Sala del Maggior Consiglio, with Tintoretto's "Paradiso" at one end; historical pictures all around; the portraits of the Doges above; a gorgeous ceiling which, I fear, demands attention; and, mercifully, the little balcony over the lagoon for escape and recovery. But first let us peep into the room on the left, where the remains of Guariento's fresco of Paradise, which Tintoretto was to supersede, have been set up: a necessarily somewhat meaningless assemblage of delicate tints and pure drawing. Then the photograph stall, which is in that ancient room of the palace that has the two beautiful windows on a lower level than the rest.
It is melancholy to look round this gigantic sala of the great Council and think of the pictures which were destroyed by the great fire in 1576, when Sebastiano Venier was Doge, among them that rendering of the battle of Lepanto, the Doge's own victory, which Tintoretto painted with such enthusiasm. A list of only a few of the works of art which from time to time have fallen to the flames would be tragic reading. Among the artists whose paintings were lost in the 1576 fire were, in addition to Tintoretto, Titian, Giovanni and Gentile Bellini, Gentile da Fabriano and Carpaccio. Sad, too, to think that the Senators who once thronged here—those grave, astute gentlemen in furred cloaks whom Tintoretto and Titian and Moroni and Moretto painted for us—assemble here no more. Sightseers now claim the palace, and the administrators of Venetian affairs meet in the Municipio, or Town Hall, on the Grand Canal.
The best thing about the room is the room itself: the courage of it in a little place like Venice! Next, I suppose, all eyes turn to the "Paradiso," and they can do nothing else if the custodian has made himself one of the party, as he is apt to do. The custodians of Venice are in the main silent, pessimistic men. They themselves neither take interest in art nor understand why you should. Their attitude to you is if not contempt only one remove from it. But one of the officials in the Doges' Palace who is sometimes to be found in this Great Hall is both enthusiastic and vocal. He has English too, a little. His weakness for the "Paradiso" is chiefly due to the circumstance that it is the "largest oil painting in the world." I dare say this is true; but the same claim, I recall, was once made for an original poster in the Strand. The "Paradiso" was one of Tintoretto's last works, the commission coming to him only by the accident of Veronese's death. Veronese was the artist first chosen, with a Bassano to assist, but when he died, Tintoretto, who had been passed over as too old, was permitted to try. The great man, painting on canvas, at the Misericordia, which had been turned into a studio for him, and being assisted by his son Domenico, finished it in 1590; and it was the delight of Venice. At first he refused payment for it, and then consented to take a present, but a smaller one than the Senate wished to offer.
The scheme of the work is logical and again illustrates his thoughtful thoroughness. At the head of all is Christ with His Mother, about and around them the angelic host led by the archangels—Michael with the scales, Gabriel with lilies, and Raphael, in prayer, each of whom presides, as we have seen, over one corner of the Palace. The next circle contains the greatest Biblical figures, Moses, David, Abraham, Solomon, Noah, the Evangelists (S. Mark prominent with his lion), and the Early Fathers. The rest of the picture is given to saints and martyrs. Not the least interesting figure is the S. Christopher, on the right, low down by the door. At his feet is the painter's daughter, for years his constant companion, who died while he was at work upon this masterpiece.
The ceiling should be examined, if one has the strength, for Veronese's sumptuous allegory of the Apotheosis of Venice. In this work the painter's wife sat for Venice, as she sat also for Europa in the picture which we have just seen in the Ante-Collegio.
On the walls are one-and-twenty representations of scenes in Venetian history devoted to the exploits of the two Doges, Sebastiano Ziani (1172-1178) and Enrico Dandolo (1192-1205). The greatest moment in the career of Ziani was the meeting of Barbarossa and the Pope, Alexander III, at S. Mark's, which has already been described; but his reign was eventful throughout. His first act as Doge was to punish the assassination of his predecessor, Vitale Michiel, who, for what was held to be the bad management of an Eastern campaign which utterly and disastrously failed, and for other reasons, was killed by the mob outside S. Zaccaria. To him succeeded Ziani and the close of the long feud between the Pope and the Emperor. It was the Pope's sojourn in Venice and his pleasure in the Venetians' hospitality which led to the elaboration of the ceremony of espousing the Adriatic. The Pope gave Ziani a consecrated ring with which to wed his bride, and much splendour was added to the pageant; while Ziani, on his return from a visit to the Pope at the Vatican, where the reconciliation with Barbarossa made it possible for the Pontiff to be at ease again, brought with him various pompous insignia that enormously increased his prestige among simple folk. It was also Ziani who had the columns of S. Theodore and the Lion erected on the Molo, while it was in his reign that the first Rialto bridge was begun. Having been Doge for six years, he retired to the monastery of S. Giorgio and there died some years later, leaving a large fortune to the poor of Venice and the church of S. Mark.
The paintings represent the Pope Alexander III recognized by the Doge when hiding in Venice; the departure of the Papal and Venetian Ambassadors for Pavia to interview the Emperor; the Pope presenting the Doge with a blessed candle; the Ambassadors before the Emperor (by Tintoretto); the Pope presenting the Doge with a sword, on the Molo; the Pope blessing the Doge; the naval battle of Salvatore, in which the Emperor Otto was captured; the Doge presenting Otto to the Pope; the Pope giving Otto his liberty; the Emperor at the Pope's feet in the vestibule of S. Mark's; the arrival of the Pope elsewhere; the Emperor and the Doge at Ancona; the Pope presenting the Doge with gifts in Rome.
Ziani seems to have been a man of address, but the great Enrico Dandolo was something more. He was a superb adventurer. He became Doge in 1193, at the trifling age of eighty-four, with eyes that had long been dimmed, and at once plunged into enterprises which, if not greatly to the good of Venice, proved his own indomitable spirit and resource. It was the time of the Fourth Crusade and the Venetians were asked to supply transports for the French warriors of the Cross to the theatre of war. After much discussion Dandolo replied that they would do so, the terms being that the Venetian vessels should carry 4500 horses, 9000 esquires, and 20,000 foot soldiers, with provisions for nine months, and for this they should be paid 85,000 silver marks. Venice also would participate in the actual fighting to the extent of providing fifty galleys, on condition that half of every conquest, whether by sea or by land, should be hers. Such was the arrangement, and the shipbuilding began at once.
But disaster after disaster occurred. The Christian commander sickened and died; a number of Crusaders backed out; others went direct to Palestine. This meant that the Venetians, who had prepared for a mighty host, incurred immense expenses which could not be met. As some reparation it was suggested to the small army of Crusaders who did arrive in the city for deportation that on their way to the Holy Land they should stop at Zara, on the Dalmatian coast, an unruly dependence of the Republic, and assist in chastising it. The objections to this course were grave. One was that the King of Hungary, in whose dominions was Zara, was a Christian and a Crusader himself; another that the Pope (Innocent III) forbade the project. Old blind Dandolo, however, was adamant. Not only must the Crusaders help the Venetians whom they had so much embarrassed by their broken bond, but he would go too. Calling the people together in S. Mark's, this ancient sightless bravo asked if it was not right that he should depart on this high mission, and they answered yes. Descending from the pulpit, he knelt at the altar and on his bonnet the Cross was fastened.
Before the expedition left, a messenger came from Alexius, nephew of the usurping King of Constantinople and son of the rightful king, praying the Venetians to sail first for Constantinople and support his father's case, and to deal faithfully with Zara later; but Dandolo said that the rebellious Zara had prior claims, and in spite of Papal threats and even excommunication, he sailed for that place on November 10, 1202. It did not take long to subdue the garrison, but winter setting in, Dandolo decided to encamp there until the spring. The delay was not profitable to the Holy Cause. The French and the Venetians grew quarrelsome, and letters from the Pope warned the French (who held him in a dread not shared by their allies) that they must leave Zara and proceed with the Crusade instantly, or expect to suffer his wrath.
Then arrived the Prince Alexius once more, with definite promises of money and men for the Crusades if the allies would come at once and win back for him the Constantinople throne. Dandolo, who saw immense Venetian advantage here, agreed, and carrying with it most of the French, the fleet sailed for the Golden Horn. Dandolo, I might remark, was now ninety-four, and it should not be forgotten that it was when he was an emissary of the Republic at Constantinople years before that he had been deprived forcibly of his sight. He was a soldier, a statesman, and (as all good Doges were) a merchant, but he was humanly mindful of past injustices too. Hence perhaps much of his eagerness to turn aside for Byzantium.
The plan was for the French to attack on the land; the Venetians on the sea. Blind though he had become, Dandolo's memory of the harbour and fortifications enabled him to arrange the naval attack with the greatest skill, and he carried all before him, himself standing on the prow of a vessel waving the banner of S. Mark. The French on land had a less rapid victory, but they won, none the less, and the ex-king Isaac was liberated and crowned once more, with his son. Both, however, instantly took to tyranny and luxurious excess, and when the time came for the promises of reward to be fulfilled nothing was done. This led to the mortification and anger of the allies, who declared that unless they were paid they would take Constantinople for themselves. War was inevitable. Meanwhile the Greeks, hating alike Venetians, French, and the Pope, proclaimed a new king, who at once killed Alexius; and the allies prepared for battle by signing a treaty, drawn up by the wily nonagenarian, in which in the event of victory Venice took literally the lion's share of the spoils.
The fighting then began. At first the Greeks were too strong, and a feeling grew among the allies that withdrawal was best; but Dandolo refused; they fought on, and Constantinople was theirs. Unhappily the victors then lost all control, and every kind of horror followed, including the wanton destruction of works of art beautiful beyond dreams. Such visible trophies of the conquest as were saved and brought back to Venice are now to be seen in S. Mark's. The four bronze horses were Dandolo's spoils, the Pala d'oro, probably the four carved columns of the high altar, and countless stone pillars and ornaments that have been worked into the structure.
The terms of the treaty were carried out faithfully, and the French paid the Venetians their original debt. Baldwin, Count of Flanders, the head of the Crusade, was named Emperor and crowned; Venice acquired large tracts of land, including the Ionian Islands; and Dandolo became "Doge of Venice, Dalmatia, and Croatia, and Lord of one-fourth and one-eighth of the Roman Empire."
The painters have chosen from Dandolo's career the following scenes: Dandolo and the Crusaders pledging themselves in S. Mark's; the capture of Zara; the request of Alexius for help; the first capture of Constantinople by Dandolo, who set the banner on the wall; the second capture of Constantinople; the election of Baldwin as Emperor; the crowning of Baldwin by Dandolo.
I said at the beginning of this precis of a gigantic campaign that it was not of great profit to Venice; nor was it. All her life she had better have listened to the Little Venice party, but particularly then, for only misfortune resulted. Dandolo, however, remains a terrific figure. He died in Constantinople in 1205 and was buried in S. Sofia. Doge Andrea Dandolo, whose tomb we saw in the Baptistery, was a descendant who came to the throne some hundred and forty years later.
Mention of Andrea Dandolo brings us to the portraits of Doges around the walls of this great hall, where the other Dandolo will also be found; for in the place adjoining Andrea's head is a black square. Once the portrait of the Doge who succeeded Andrea was here too, but it was blacked out. Marino Faliero, for he it was, became Doge in 1354 when his age was seventy-six, having been both a soldier and a diplomatist. He found himself at once involved in the war with Genoa, and almost immediately came the battle of Sapienza, when the Genoese took five thousand prisoners, including the admiral, Niccolo Pisani. This blow was a very serious one for the Venetians, involving as it did great loss of life, and there was a growing feeling that they were badly governed. The Doge, who was but a figure-head of the Council of Ten, secretly thinking so too, plotted for the overthrow of the Council and the establishment of himself in supreme power. The Arsenal men were to form his chief army in the revolt; the false alarm of a Genoese attack was to get the populace together; and then the blow was to be struck and Faliero proclaimed prince. But the plot miscarried through one of the conspirators warning a friend to keep indoors; the ringleaders were caught and hanged or exiled; and the Doge, after confessing his guilt, was beheaded in the courtyard of this palace. His coffin may be seen in the Museo Civico, and of his unhappy story Byron made a drama.
One of Faliero's party was Calendario, an architect, employed on the part of the Doges' Palace in which we are now standing. He was hanged or strangled between the two red columns in the upper arches of the Piazzetta facade.
The first Doge to be represented here is Antenorio Obelerio (804-810), but he had had predecessors, the first in fact dating from 697. Of Obelerio little good is known. He married a foreigner whom some believe to have been an illegitimate daughter of Charlemagne, and her influence was bad. His brother Beato shared his throne, and in the end probably chased him from it. Beato was Doge when Rialto became the seat of government, Malamocco having gone over to the Franks under Pepin. But of Beato no account is here taken, Obelerio's successor being Angelo Partecipazio (810-827), who was also the first occupant of the first Ducal Palace, on the site of a portion of the present one. It was his son Giustiniano, sharing the throne with his father, who hit upon the brilliant idea of stealing the body of S. Mark from Alexandria and of preserving it in Venice, thus establishing that city not only as a religious centre but also as a place of pilgrimage and renown. As Mrs. Richardson remarks in her admirable survey of the Doges: "Was it not well that the government of the Doge Giustiniano and his successors throughout the age should become the special concern of a Saint-Evangelist in whose name all national acts might be undertaken and accomplished; all national desires and plans—as distinct from and dominant over purely ecclesiastical ones—be sanctified and made righteous?" The success of the scheme of theft I have related in an earlier chapter; and how this foresight was justified, history tells. It is odd that Venice does not make more acclamation of Giustiniano (or Partecipazio II). To his brother Giovanni, who early had shown regrettable sympathy with the Franks and had been banished accordingly, Giustiniano bequeathed the Dogeship (as was then possible), and it was in his reign (829-836) that S. Mark's was begun.
The last Doge in this room is Girolamo Priuli (1559-1567), of whom nothing of account is remembered save that it was he who invited Tintoretto to work in the palace and on one of the ceilings. You may see his portrait in one of the rooms, from Tintoretto's brush, in the company of Venice, Justice, S. Mark and the Lion.
Of the others of the six-and-seventy Doges around the room I do not here speak. The names of such as are important will be found elsewhere throughout this book, as we stand beside their tombs or glide past their palaces.
Before leaving the Hall one should, as I have said, walk to the balcony, the door of which the custodian opens for each visitor with a mercenary hand. It should of course be free to all; and Venice would do well to appoint some official (if such could be found) to enforce such liberties. Immediately below is all the movement of the Molo; then the edge of the lagoon with its myriad gondolas; then the sparkling water, with all its busy activities and swaying gondoliers; and away beyond it the lovely island of S. Giorgio. A fairer prospect the earth cannot show.
The first Doge in the Sala dello Scrutinio is Pietro Loredan (1567-1570) and the last of all Lodovico Manin (1788-1797) who fell before the inroads of Napoleon. "Take it away," he said to his servant, handing him the linen cap worn beneath the ducal corno, "we shall not need it any more." He retired into piety and left his fortune to good works.
This room, also a fine and spacious hall but smaller than the Sala del Maggior Consiglio, has historical pictures, and a "Last Judgment," by Palma the younger, which immensely interests the custodian by reason of a little human touch which may or may not be true. On the left of the picture, in the Infernal regions, low down, will be seen a large semi-nude female sinner in torment; on the right, in heaven, the same person is seen again, in bliss. According to the custodian this lady was the painter's innamorata, and he set her in both places as a reward for her varying moods. The other pictures represent the capture of Zara by Marco Giustiniani in 1346. Zara, I may mention, had very badly the habit of capture: this was the eighth time it had fallen. Tintoretto is the painter, and it is one of his best historical works. The great sea-fight picture on the right wall represents another battle of Lepanto, a later engagement than Venier's; the painter is Andrea Vicentino, who has depicted himself as the figure in the water; while in another naval battle scene, in the Dardanelles, the painter, Pietro Liberi, is the fat naked slave with a poniard. For the rest the guide-book should be consulted. The balcony of the room, which juts over the Piazzetta, is rarely accessible; but if it is open one should tarry there for the fine view of Sansovino's Old Library.
The second set of showrooms (which require the expenditure of another lira)—the oldest rooms in the palace—constitute the Archaeological Museum. Here one sees a few pictures, a few articles of vertu, some sumptuous apartments, some rich ceilings, and a wilderness of ancient sculpture. The first room shown, the Sala degli Scarlatti, is the bedroom of the Doges, with a massive and rather fine chimney piece and an ornate ceiling. The next room, the Sala dello Scudo, has a fine decorative, if inaccurate, map of the world, made by a monk in the fifteenth century. The next, the Sala Grimani, has rival lions of S. Mark by Jacobello del Fiore, an early Venetian painter, in 1415, and Carpaccio a century later. Jacopo's lion has a very human face; Carpaccio's picture is finer and is also interesting for its architectural details. The next room, the Sala Erizzo, has a very splendid ceiling. The next is not remarkable, and then we come on the right to the Sala dei Filosofi where the custodian displays, at the foot of the staircase, the charming fresco of S. Christopher which Titian made for Doge Andrea Gritti. It is a very pleasing rendering, and the Christ Child never rode more gaily or trustfully on the friendly saint. With true patriotism Titian has placed the incident in a shallow of the lagoon and the Doges' Palace is seen in the distance.
Then follow three rooms in the Doges' suite in which a variety of treasures are preserved, too numerous and heterogeneous for description.
The antique section of the Archaeological Museum is not of general interest. It consists chiefly of Greek and Roman sculpture collected by Cardinal Grimani or dug from time to time from the soil of Venetian provinces. Here are a few beautiful or precious relics and much that is indifferent. In the absence of a Hermaphrodite, the most popular possession is (as ever) a group of Leda and the Swan. I noted among the more attractive pieces a Roman altar with lovers (Baedeker calls them satyrs), No. 68; a Livia in black marble, No. 102; a nice girl, Giulia Mammea, No. 142; a boy, very like a Venetian boy of to-day, No. 145; a giant Minerva, No. 169; a Venus, No. 174; an Apollo, No. 223. A very beautiful Pieta by Giovanni Bellini, painted under the influence of Duerer, should be sought and found.
The Bridge of Sighs, a little way upon which one may venture, is more interesting in romantic fancy than in fact, and its chief merit is to span very gracefully the gulf between the Palace and the Prison. With the terrible cells of the Doges' Palace, to which we are about to descend, it has no connexion. When Byron says, in the famous line beginning the fourth canto of "Childe Harold,"
I stood in Venice on the Bridge of Sighs,
he probably meant that he stood in Venice on the Bridge of Straw (Ponte di Paglia) and contemplated the Bridge of Sighs. Because one does not stand on the Bridge of Sighs but in it, for it is merely dark passages lit by gratings. But to stand on the Ponte di Paglia on the Riva and gaze up the sombre Rio del Palazzo with the famous arch poised high over it is one of the first duties of all visitors to Venice and a very memorable experience.
Lastly, the horrible cells (which cost half a lira more), upon which and the damp sinister rooms where the place of execution and oubliette were situated, a saturnine custodian says all that is necessary. Let me, however, quote a warning from the little Venetian guide-book: "Everybody to whom are pointed out the prisons to which Carmagnola, Jacopo Foscari, Antonio Foscarini, etc., were confined, will easily understand that such indications cannot be true at all."
THE DOGES' PALACE. II: THE EXTERIOR
The colour of Venice—Sunny Gothic—A magical edifice—The evolution of a palace—A fascinating balcony—The carved capitals—A responsible column—The Porta della Carta—The lions of Venice—The Giants' Stairs—Antonio Rizzo—A closed arcade—Casanova—The bronze wells—A wonderful courtyard—Anonymous accusations—A Venetian Valhalla.
"That house," said an American on a Lido steamboat, pointing to the Doges' Palace, "is a wonder in its way."
Its way is unique. The soft gentle pink of its south and west facades remains in the memory as long and as firmly as the kaleidoscopic hues of S. Mark's. This pink is, I believe, the colour of Venice.
Whether or not the Doges' Palace as seen from S. Giorgio Maggiore, with its seventeen massive arches below, its thirty-four slender arches above, above them its row of quatrefoiled circles, and above them its patterned pink wall with its little balcony and fine windows, the whole surmounted by a gay fringe of dazzling white stone—whether or not this is the most beautiful building in the world is a question for individual decision; but it would, I think, puzzle anyone to name a more beautiful one, or one half so charming. There is nothing within it so entrancing as its exterior—always with the exception of Tintoretto's, "Bacchus and Ariadne."
The Ducal Palace is Gothic made sprightly and sunny; Gothic without a hint of solidity or gloom. So light and fresh is the effect, chiefly the result of the double row of arches and especially of the upper row, but not a little due to the zig-zagging of the brickwork and the vivid cheerfulness of the coping fringe, that one has difficulty in believing that the palace is of any age at all or that it will really be there to-morrow. The other buildings in the neighbourhood—the Prison, the Mint, the Library, the Campanile: these are rooted. But the Doges' Palace might float away at any moment. Aladdin's lamp set it there: another rub and why should it not vanish?
The palace as we see it now has been in existence from the middle of the sixteenth century. Certain internal changes and rebuildings have occurred, but its facades on the Piazzetta and lagoon, the Giants' Stairs, the courtyard, were then as now. But before that time constant structural modification was in progress. The original palace ran beside the Rio del Palazzo from S. Mark's towers to the Ponte di Paglia, with a wing along the lagoon. Its width was equal to that from the present Noah or Vine Corner by the Ponte di Paglia to the fifth column from that corner. Its wing extended to the Piazzetta. A wall and moat protected it, the extent of its ramparts being practically identical with the extent of the present building. This, the first, palace was erected in the ninth century, after the seat of government was changed from Malamocco to Venice proper.
Various conflagrations, in addition to the growing needs of the State, led to rebuilding and enlargement. The first wing was added in the twelfth century, when the basement and first floor of the portion from the Porta della Carta to the thick seventh column from the Adam and Eve group, under the medallion of Venice, on the Piazzetta facade, was set up, but not in the style which we now know. That was copied three centuries later from the Riva or lagoon facade. In 1301 the hall above the original portion on the Rio del Palazzo side, now called the Sala del Senato, was added and the lagoon wing was rebuilt, the lower arches, which are there to-day, being then established. A few years later, a still greater hall being needed, the present Sala del Maggior Consiglio was erected, and this was ready for use in 1423. The lagoon facade as we see it now, with its slender arches above the sturdy arches, thus dates from the beginning of the fifteenth century, and this design gave the key to the builders of later Venice, as a voyage of the Grand Canal will prove.
It was the great Doge Tommaso Mocenigo (1413-1423) who urged upon the Senate the necessity of completing the palace. In 1424 the work was begun. Progress was slow and was hindered by the usual fire, but gradually the splendid stone wall on the Rio del Palazzo side went up, and the right end of the lagoon facade, and the Giants' Stairs, and the Piazzetta facade, reproducing the lagoon facade. The elaborately decorated facades of the courtyard came later, and by 1550 the palace was finished. The irregularity of the windows on the lagoon facade is explained by this piecemeal structure. The four plain windows and the very graceful balcony belong to the Sala del Maggior Consiglio. The two ornate windows on the right were added when the palace was brought into line with this portion, and they are lower because the room they light is on a level lower than the great Council Hall's. The two ugly little square windows (Bonington in his picture in the Louvre makes them three) probably also were added then.
When the elegant spired cupolas at each corner of the palace roof were built, I do not know, but they look like a happy afterthought. The small balcony overlooking the lagoon, which is gained from the Sala del Maggior Consiglio, and which in Canaletto and Guardi's eighteenth-century pictures always, as now, has a few people on it, was built in 1404. It is to be seen rightly only from the water or through glasses. The Madonna in the circle is charming. She has one child in her arms and two at her knees, and her lap is a favourite resting-place for pigeons. In the morning when the day is fine the green bronze of the sword and crown of Justice (or, as some say, Mars), who surmounts all, is beautiful against the blue of the sky.
The Piazzetta facade balcony was built early in the sixteenth century, but the statue of S. George is a recent addition, Canova being the sculptor.
Now let us examine the carved capitals of the columns of the Ducal Palace arcade, for these are extremely interesting and transform it into something like an encyclopedia in stone. Much thought has gone to them, the old Venetians' love of symbols being gratified often to our perplexity. We will begin at the end by the Porta della Carta, under the group representing the Judgment of Solomon—the Venetians' platonic affection for the idea of Justice being here again displayed. This group, though primitive, the work of two sculptors from Fiesole early in the fifteenth century, has a beauty of its own which grows increasingly attractive as one returns and returns to the Piazzetta. Above the group is the Angel Gabriel; below it, on the richly foliated capital of this sturdy corner column, which bears so much weight and splendour, is Justice herself, facing Sansovino's Loggetta: a little stone lady with scales and sword of bronze. Here also is Aristotle giving the law to some bearded men; while other figures represent Solon, another jurist, Scipio the chaste, Numa Pompilius building a church, Moses receiving the tables of the law, and Trajan on horseback administering justice to a widow. All are named in Latin.
The second capital has cherubs with fruit and birds and no lettering.
The third has cranes and no lettering.
The fourth is allegorical, representing, but without much psychology, named virtues and vices, such as misery, cheerfulness, folly, chastity, honesty, falsehood, injustice, abstinence.
The fifth has figures and no lettering. A cobbler faces the campanile. It is above this fifth column that we notice in the upper row of arches two columns of reddish stain. It was between these that malefactors were strangled.
The sixth has symbolical figures which I do not understand. Ruskin suggests that they typify the degradation of human instincts. A knight in armour is here. A musician seated on a fish faces the Old Library. There is no lettering, and as is the case throughout the figures on the wall side are difficult to discern.
The seventh represents the vices, and names them: luxury, gluttony, pride, anger, avarice, idleness, vanity, envy.
The eighth represents the virtues and names them: hope, faith, fortitude, temperance, humility, charity, justice, prudence.
The ninth has virtues and vices, named and mixed: modesty, discord, patience, constancy, infidelity, despair, obedience, liberality.
The tenth has named fruits.
Ruskin thinks that the eleventh may illustrate various phases of idleness. It has no lettering.
The twelfth has the months and their employments, divided thus: January (indoors) and February, March blowing his pipes, April with a lamb and May, June (the month of cherries), July with a sheaf of corn and August, September (the vintage), October and November, and December, pig-sticking.
The thirteenth, on a stouter column than the others, because it has a heavier duty, namely, to bear the party wall of the great Council Hall, depicts the life of man. There is no lettering. The scenes represent love (apparently at first sight), courtship, the marriage bed, and so forth, the birth of the baby, his growth and his death. Many years ago this column was shown to me by the captain of a tramp steamer, as the most interesting thing in Venice; and there are others who share his opinion. Above it on the facade is the medallion of the Queen of the Adriatic ruling her domains.
The fourteenth capital represents national types, named: Persian, Latin, Tartar, Turk, Hungarian, Greek, Goth, and Egyptian.
The fifteenth is more elaborate and ingenious. It represents the ages of man and his place in the stellar system. Thus, infancy is governed by the moon, childhood by Mercury, youth by the sun, and so forth.
The sixteenth depicts various craftsmen: the smith, the mason, the goldsmith, the carpenter, the notary, the cobbler, the man-servant, the husbandman. Over this are traces of a medallion, probably of porphyry, now removed.
The seventeenth has the heads of animals: lion, bear, wolf, and so forth, including the griffin each with its prey.
The eighteenth has eight stone-carving saints, some with a piece of coloured marble, all named, and all at work: S. Simplicius, S. Symphorian, who sculps a figure, S. Claudius, and others.
And now we are at the brave corner column which unconcernedly assumes a responsibility that can hardly be surpassed in the world. For if it were to falter all would go. Down would topple two of the loveliest facades that man ever constructed or the centuries ever caressed into greater beauty. This corner of the palace has an ever-increasing fascination for me, and at all hours of the day and night this strong column below and the slenderer one above it hold the light—whether of sun or moon or artifice—with a peculiar grace.
The design of this capital is, fittingly enough, cosmic. It represents the signs of the Zodiac with the addition, on the facet opposite the Dogana, of Christ blessing a child. Facing S. Giorgio are Aquarius and Capricornus, facing the Lido are Pisces and Sagittarius. Elsewhere are Justice on the Bull, the Moon in a boat with a Crab, and a Virgin reading to the Twins.
Above this capital, on the corner of the building itself, are the famous Adam and Eve, presiding over the keystone of the structure as over the human race. It is a naive group, as the photograph shows, beneath the most tactful of trees, and it has no details of beauty; and yet, like its companions, the Judgment of Solomon and the Sin of Ham, it has a curious charm—due not a little perhaps to the softening effect of the winds and the rains. High above our first parents is the Angel Michael.
The first capital after the corner (we are now proceeding down the Riva) has Tubal Cain the musician, Solomon, Priscian the grammarian, Aristotle the logician, Euclid the geometrician, and so forth, all named and all characteristically employed.
The second has heads of, I suppose, types. Ruskin suggests that the best looking is a Venetian and the others the Venetians' inferiors drawn from the rest of the world.
The third has youths and women with symbols, signifying I know not what. All are corpulent enough to suggest gluttony. This is repeated in No. 11 on the Piazzetta side.
The fourth has various animals and no lettering.
The fifth has lions' heads and no lettering.
The sixth has virtues and vices and is repeated in the fourth on the Piazzetta.
The seventh has cranes, and is repeated in the third on the Piazzetta.
The eighth has vices again and is repeated in the seventh on the Piazzetta. Above it are traces of a medallion over three triangles.
The ninth has virtues and is repeated in the eighth on the Piazzetta.
The tenth has symbolical figures, and is repeated in the sixth on the Piazzetta.
The eleventh has vices and virtues and is repeated in the ninth on the Piazzetta.
The twelfth has female heads and no lettering.
The thirteenth has named rulers: Octavius, Titus, Trajan, Priam, Darius, and so forth, all crowned and ruling.
The fourteenth has children and no lettering.
The fifteenth has heads, male and female, and no lettering. Above it was once another medallion and three triangles.
The sixteenth has pelicans and no lettering.
The seventeenth and last has children with symbols and no lettering.
Above this, on the corner by the bridge, is the group representing the Sin of Ham. Noah's two sons are very attractive figures. Above the Noah group is the Angel Raphael.
The gateway of the palace—the Porta della Carta—was designed by Giovanni and Bartolommeo Bon, father and son, in the fourteen thirties and forties. Francesco Foscari (1423-1457) being then Doge, it is he who kneels to the lion on the relief above, and again on the balcony of the Piazzetta facade. At the summit of the portal is Justice once more, with two attendant lions, cherubs climbing to her, and live pigeons for ever nestling among them. I counted thirty-five lions' heads in the border of the window and thirty-five in the border of the door, and these, with Foscari's one and Justice's two, and those on the shields on each side of the window, make seventy-five lions for this gateway alone. Then there are lions' heads between the circular upper arches all along each facade of the palace.
It would be amusing to have an exact census of the lions of Venice, both winged and without wings. On the Grand Canal alone there must be a hundred of the little pensive watchers that sit on the balustrades peering down. As to which is the best lion, opinions must, of course, differ, the range being so vast: between, say, the lion on the Molo column and Daniele Manin's flamboyant sentinel at the foot of the statue in his Campo. Some would choose Carpaccio's painted lion in this palace; others might say that the lion over the Giants' Stairs is as satisfying as any; others might prefer that fine one on the Palazzo dei Camerlenghi by the Rialto bridge, and the Merceria clock tower's lion would not want adherents.
Why this lovely gateway was called the Porta della Carta (paper) is not absolutely certain: perhaps because public notices were fixed to its door; perhaps because paper-sellers frequented it; perhaps because the scriveners of the Republic worked hereabouts. Passing through it we have before us the Giants' Stairs, designed by Antonio Rizzo and taking their name from the two great figures of Mars and Neptune at the top by Jacopo Sansovino. On the upright of each step is a delicate inlaid pattern—where, in England, so often we read of the virtues of malted milk or other commodity. Looking back from the foot of the stairs we see Sansovino's Loggetta, framed by the door; looking back from the top of the stairs we have in front of us Rizzo's statues of Adam and Eve. This Antonio Rizzo, or Ricci, who so ably fortified Sansovino as a beautifier of Venice, was a Veronese, of whom little is known. He flourished in the second half of the fifteenth century.
Every opportunity of passing through the courtyard should be taken, and during the chief hours of the day there is often—but not invariably—a right of way between the Porta della Carta and the Riva, across the courtyard, while the first floor gallery around it, gained by the Giants' Stairs, is also open. For one of those capricious reasons, of which Italian custodians everywhere hold the secret, the delightful gallery looking on the lagoon and Piazzetta is, however, closed. I once found my way there, but was pursued by a frantic official and scolded back again.
The courtyard is inexhaustible in interest and beauty, from its bronze well-heads to the grated leaden prison cells on the roof, the terrible piombi which were so dreaded on account of their heat in summer and cold in winter. Here in the middle of the eighteenth century that diverting blackguard, Jacques Casanova, was imprisoned. He was "under the leads" over the Piazzetta wing, and the account of his durance and his escape is one of the most interesting parts, and certainly the least improper, of his remarkably frank autobiography. Venice does not seem to have any pride in this son of hers, but as a master of licentiousness, effrontery, adventurousness, and unblushing candour he stands alone in the world. Born at Venice in 1725, it was in the seminary of S. Cyprian here that he was acquiring the education of a priest when events occurred which made his expulsion necessary. For the history of his utterly unprincipled but vivacious career one must seek his scandalous and diverting pages. In 1755, on an ill-starred return visit to his native city, he was thrown into this prison, but escaping and finding his way to Paris, he acquired wealth and position as the Director of State Lotteries. Casanova died in 1798, but his memories cease with 1774. His pages may be said to supply a gloss to Longhi's paintings, and the two men together complete the picture of Venetian frivolity in their day and night.
The well-head nearer the Giants' Stairs was the work of Alberghetti and is signed inside. The other has the head of Doge Francesco Venier (1554-1556) repeated in the design and is stated within to be the work of Niccolo Conti, a son of Venice. Coryat has a passage about the wells which shows how much more animated a scene the ducal courtyard used to present than now. "They yeeld very pleasant water," he writes. "For I tasted it. For which cause it is so much frequented in the Sommer time that a man can hardly come thither at any time in the afternoone, if the sunne shineth very hote, but he shall finde some company drawing of water to drinke for the cooling of themselves." To-day they give water no more, nor do the pigeons come much to the little drinking place in the pavement here but go rather to that larger one opposite Cook's office.
Everything that an architect can need to know—and more—may be learned in this courtyard, which would be yet more wonderful if it had not its two brick walls. Many styles meet and mingle here: Gothic and Renaissance, stately and fanciful, sombre and gay. Every capital is different. Round arches are here and pointed; invented patterns and marble with symmetrical natural veining which is perhaps more beautiful. Every inch has been thought out and worked upon with devotion and the highest technical skill; and the antiseptic air of Venice and cleansing sun have preserved its details as though it were under glass.
In the walls beneath the arcade on the Piazzetta side may be seen various ancient letter-boxes for the reception of those accusations against citizens, usually anonymous, in which the Venetians seem ever to have rejoiced. One is for charges of evading taxation, another for those who adulterate bread, and so forth.
The upper gallery running round the courtyard has been converted into a Venetian—almost an Italian—Valhalla. Here are busts of the greatest men, and of one woman, Catherine Cornaro, who gave Cyprus to the Republic and whom Titian painted. Among the first busts that I noted—ascending the stairs close to the Porta della Carta—was that of Ugo Foscolo, the poet, patriot, and miscellaneous writer, who spent the last years of his life in London and became a contributor to English periodicals. One of his most popular works in Italy was his translation of Sterne's Sentimental Journey. He died at Turnham Green in 1827, but his remains, many years after, were moved to Santa Croce in Florence. Others are Carlo Zeno, the soldier; Goldoni, the dramatist; Paolo Sarpi, the monkish diplomatist; Galileo Galilei, the astronomer and mathematician; the two Cabots, the explorers, and Marco Polo, their predecessor; Niccolo Tommaseo, the patriot and associate of Daniele Manin, looking very like a blend of Walt Whitman and Tennyson; Dante; a small selection of Doges, of whom the great Andrea Dandolo is the most striking; Tintoretto, Giovanni Bellini, Titian, and Paul Veronese; Tiepolo, a big-faced man in a wig whom the inscription credits with having "renewed the glory" of the two last named; Canova, the sculptor; Daniele Manin, rather like John Bright; Lazzaro Mocenigo, commander in chief of the Venetian forces, rather like Buffalo Bill; and flanking the entrance to the palace Vittorio Pisani and Carlo Zeno, the two patriots and warriors who together saved the Republic in the Chioggian war with the Genoese in the fourteenth century.
This collection of great men makes no effort to be complete, but it is rather surprising not to find such very loyal sons of Venice as Canaletto, Guardi and Longhi among the artists, and Giorgione is of course a grievous omission.
THE PIAZZETTA
The two columns—An ingenious engineer—S. Mark's lion—S. Theodore of Heraclea—The Old Library—Jacopo Sansovino—The Venetian Brunelleschi—Vasari's life—A Venetian library—Early printed books—The Grimani breviary—A pageant of the Seasons—The Loggetta—Coryat again—The view from the Molo—The gondolier—Alessandro and Ferdinando—The danger of the traghetto—Indomitable talkers—The fair and the fare—A proud father—The rampino.
The Piazzetta is more remarkable in its architectural riches than the Piazza. S. Mark's main facade is of course beyond words wonderful; but after this the Piazza has only the Merceria clock and the Old and the New Procuratie, whereas the Piazzetta has S. Mark's small facade, the Porta della Carta and lovely west facade of the Doges' Palace, the columns bearing S. Mark's lion and S. Theodore, Sansovino's Old Library and Loggetta; while the Campanile is common to both. The Piazzetta has a cafe too, although it is not on an equality either with Florian's or the Quadri, and on three nights a week a band plays.
The famous Piazzetta columns, with S. Theodore and his crocodile (or dragon) on one and the lion of S. Mark on the other, which have become as much a symbol of Venice as the facade of S. Mark's itself, were brought from Syria after the conquest of Tyre. Three were brought in all, but one fell into the water and was never recovered. The others lay on the quay here for half a century waiting to be set up, a task beyond human skill until an engineer from Lombardy volunteered to do it on condition that he was to have any request granted. His request was to be allowed the right of establishing a gaming-table between the columns; and the authorities had to comply, although gambling was hateful to them. A few centuries later the gallows were placed here too. Now there is neither gambling nor hanging; but all day long loafers sit on the steps of the columns and discuss pronto and subito and cinque and all the other topics of Venetian conversation.
I wonder how many visitors to Venice, asked whether S. Theodore on his column and the Lion of S. Mark on his, face the lagoon or the Merceria clock, would give the right answer. The faces of both are turned towards the clock; their backs to the lagoon. The lion, which is of bronze with white agates for his eyes, has known many vicissitudes. Where he came from originally, no one knows, but it is extremely probable that he began as a pagan and was pressed into the service of the Evangelist much later. Napoleon took him to Paris, together with the bronze horses, and while there he was broken. He came back in 1815 and was restored, and twenty years ago he was restored again. S. Theodore was also strengthened at the same time, being moved into the Doges' Palace courtyard for that purpose.
There are several saints named Theodore, but the protector and patron of the Venetians in the early days before Mark's body was stolen from Alexandria, is S. Theodore of Heraclea. S. Theodore, surnamed Stretelates, or general of the army, was a famous soldier and the governor of the country of the Mariandyni, whose capital was Heraclea. Accepting and professing the Christian faith, he was beheaded by the Emperor Licinius on February 7, 319. On June 8 in the same year his remains were translated to Euchaia, the burial-place of the family, and the town at once became so famous as a shrine that its name was changed to Theodoropolis. As late as 970 the patronage of the Saint gave the Emperor John I a victory over the Saracens, and in gratitude the emperor rebuilt the church where Theodore's relics were preserved. Subsequently they were moved to Mesembria and then to Constantinople, from which city the great Doge Dandolo brought them to Venice. They now repose in S. Salvatore beneath an altar.
The west side of the Piazzetta consists of the quiet and beautiful facade of Sansovino's Old Library. To see it properly one should sit down at ease under the Doge's arcade or mount to the quadriga gallery of S. Mark's. Its proportions seem to me perfect, but Baedeker's description of it as the most magnificent secular edifice in Italy seems odd with the Ducal Palace so near. They do not, however, conflict, for the Ducal Palace is so gay and light, and this so serious and stately. The cherubs with their garlands are a relaxation, like a smile on a grave face; yet the total effect is rather calm thoughtfulness than sternness. The living statues on the coping help to lighten the structure, and if one steps back along the Riva one sees a brilliant column of white stone—a chimney perhaps—which is another inspiriting touch. In the early morning, with the sun on them, these statues are the whitest things imaginable.
The end building, the Zecca, or mint, is also Sansovino's, as are the fascinating little Loggetta beneath the campanile, together with much of its statuary, the giants at the head of Ricco's staircase opposite, and the chancel bronzes in S. Mark's, so that altogether this is peculiarly the place to inquire into what manner of man the Brunelleschi of Venice was. For Jacopo Sansovino stands to Venice much as that great architect to Florence. He found it lacking certain essential things, and, supplying them, made it far more beautiful and impressive; and whatever he did seems inevitable and right.
Vasari wrote a very full life of Sansovino, not included among his other Lives but separately published. In this we learn that Jacopo was born in Florence in 1477, the son of a mattress-maker named Tatti; but apparently 1486 is the right date. Appreciating his natural bent towards art, his mother had him secretly taught to draw, hoping that he might become a great sculptor like Michael Angelo, and he was put as apprentice to the sculptor Andrea Contucci of Monte Sansovino, who had recently set up in Florence and was at work on two figures for San Giovanni; and Jacopo so attached himself to the older man that he became known as Sansovino too. Another of his friends as a youth was Andrea del Sarto.
From Florence he passed to Rome, where he came under the patronage of the Pope Julius II, of Bramante, the architect, and of Perugino, the painter, and learned much by his studies there. Returning to Florence, he became one of the most desired of sculptors and executed that superb modern-antique, the Bacchus in the Bargello. Taking to architecture, he continued his successful progress, chiefly again in Rome, but when the sack of that city occurred in 1527 he fled and to the great good fortune of Venice took refuge here. The Doge, Andrea Gritti, welcomed so distinguished a fugitive and at once set him to work on the restoration of S. Mark's cupolas, and this task he completed with such skill that he was made a Senior Procurator and given a fine house and salary.
As a Procurator he seems to have been tactful and active, and Vasari gives various examples of his reforming zeal by which the annual income of the Procuranzia was increased by two thousand ducats. When, however, one of the arches of Sansovino's beautiful library fell, owing to a subsidence of the foundations, neither his eminent position nor ability prevented the authorities from throwing him into prison as a bad workman; nor was he liberated, for all his powerful friends, without a heavy fine. He built also several fine palaces, the mint, and various churches, but still kept time for his early love, sculpture, as his perfect little Loggetta, and the giants on the Staircase, and such a tomb as that in S. Salvatore, show.
This is Vasari's description of the man: "Jacopo Sansovino, as to his person, was of the middle height, but rather slender than otherwise, and his carriage was remarkably upright; he was fair, with a red beard, and in his youth was of a goodly presence, wherefore he did not fail to be loved, and that by dames of no small importance. In his age he had an exceedingly venerable appearance; with his beautiful white beard, he still retained the carriage of his youth: he was strong and healthy even to his ninety-third year, and could see the smallest object, at whatever distance, without glasses, even then. When writing, he sat with his head up, not supporting himself in any manner, as it is usual for men to do. He liked to be handsomely dressed, and was singularly nice in his person. The society of ladies was acceptable to Sansovino, even to the extremity of age, and he always enjoyed conversing with or of them. He had not been particularly healthy in his youth, yet in his old age he suffered from no malady whatever, in-so-much that, for a period of fifty years, he would never consult any physician even when he did feel himself indisposed. Nay, when he was once attacked by apoplexy, he would still have nothing to do with physic, but cured himself by keeping in bed for two months in a dark and well-warmed chamber. His digestion was so good that he could eat all things without distinction: during the summer he lived almost entirely on fruits, and in the very extremity of his age would frequently eat three cucumbers and half a lemon at one time.
"With respect to the qualities of his mind, Sansovino was very prudent; he foresaw readily the coming events, and sagaciously compared the present with the past. Attentive to his duties, he shunned no labour in the fulfilment of the same, and never neglected his business for his pleasure. He spoke well and largely on such subjects as he understood, giving appropriate illustrations of his thoughts with infinite grace of manner. This rendered him acceptable to high and low alike, as well as to his own friends. In his greatest age his memory continued excellent; he remembered all the events of his childhood, and could minutely refer to the sack of Rome and all the other occurrences, fortunate or otherwise, of his youth and early manhood. He was very courageous, and delighted from his boyhood in contending with those who were greater than himself, affirming that he who struggles with the great may become greater, but he who disputes with the little must become less. He esteemed honour above all else in the world, and was so upright a man of his word, that no temptation could induce him to break it, of which he gave frequent proof to his lords, who, for that as well as other qualities, considered him rather as a father or brother than as their agent or steward, honouring in him an excellence that was no pretence, but his true nature."
Sansovino died in 1570, and he was buried at San Gimignano, in a church that he himself had built. In 1807, this church being demolished, his remains were transferred to the Seminario della Salute in Venice, where they now are.
Adjoining the Old Library is the Mint, now S. Mark's Library, which may be both seen and used by strangers. It is not exactly a British Museum Reading-room, for there are but twelve tables with six seats at each, but judging by its usually empty state, it more than suffices for the scholarly needs of Venice. Upstairs you are shown various treasures brought together by Cardinal Bessarione: MSS., autographs, illuminated books, and incunabula. A fourteenth-century Dante lies open, with coloured pictures: the poet very short on one page and very tall on the next, and Virgil, at his side, very like Christ. A Relazione della Morte de Anna Regina de Francia, a fifteenth-century work, has a curious picture of the queen's burial. The first book ever printed in Venice is here: Cicero's Epistolae, 1469, from the press of Johannes di Spira, which was followed by an edition of Pliny the Younger. A fine Venetian Hypnerotomachia, 1499, is here, and a very beautiful Herodotus with lovely type from the press of Gregorius of Venice in 1494. Old bindings may be seen too, among them a lavish Byzantine example with enamels and mosaics. The exhibited autographs include Titian's hand large and forcible; Leopardi's, very neat; Goldoni's, delicate and self-conscious; Galileo's, much in earnest; and a poem by Tasso with myriad afterthoughts.
But the one idea of the custodian is to get you to admire the famous Grimani Breviary—not alas! in the original, which is not shown, but in a coloured reproduction. Very well, you say; and then discover that the privilege of displaying it is the perquisite of a rusty old colleague. That is to say, one custodian extols the work in order that another may reap a second harvest by turning its leaves. This delightful book dates from the early sixteenth century and is the work of some ingenious and masterly Flemish miniaturist with a fine sense of the open air and the movement of the seasons. But it is hard to be put off with an ordinary bookseller's traveller's specimen instead of the real thing. If one may be so near Titian's autograph and the illuminated Divine Comedy, why not this treasure too? January reveals a rich man at his table, dining alone, with his servitors and dogs about him; February's scene is white with snow—a small farm with the wife at the spinning-wheel, seen through the door, and various indications of cold, without; March shows the revival of field labours; April, a love scene among lords and ladies; May, a courtly festival; June, haymaking outside a fascinating city; July, sheep-shearing and reaping; August, the departure for the chase; September, grape-picking for the vintage; October, sowing seeds in a field near another fascinating city—a busy scene of various activities; November, beating oak-trees to bring down acorns for the pigs; and December, a boar hunt—the death. And all most gaily coloured, with the signs of the Zodiac added.
The little building under the campanile is Sansovino's Loggetta, which he seems to have set there as a proof of his wonderful catholicity—to demonstrate that he was not only severe as in the Old Library, and Titanic as in the Giants, but that he had his gentler, sweeter thoughts too. The Loggetta was destroyed by the fall of the campanile; but it has risen from its ruins with a freshness and vivacity that are bewildering. It is possible indeed to think of its revivification as being more of a miracle than the new campanile: for the new campanile was a straight-forward building feat, whereas to reconstruct Sansovino's charm and delicacy required peculiar and very unusual gifts. Yet there it is: not what it was, of course, for the softening quality of old age has left it, yet very beautiful, and in a niche within a wonderful restoration of Sansovino's group of the Madonna and Child with S. John. The reliefs outside have been pieced together too, and though here and there a nose has gone, the effect remains admirable. The glory of Venice is the subject of all.
The most superb of the external bronzes is the "Mercury" on the left of the facade. To the patience and genius of Signor Giacomo Boni is the restored statuary of the Loggetta due; Cav. Munaretti was responsible for the bronzes, and Signor Moretti for the building. All honour to them!
Old Coryat's enthusiasm for the Loggetta is very hearty. "There is," he says, "adjoyned unto this tower [the campanile] a most glorious little roome that is very worthy to be spoken of, namely the Logetto, which is a place where some of the Procurators of Saint Markes doe use to sit in judgement, and discusse matters of controversies. This place is indeed but little, yet of that singular and incomparable beauty, being made all of Corinthian worke, that I never saw the like before for the quantity thereof."
Where the Piazzetta especially gains over the Piazza is in its lagoon view. From its shore you look directly over the water to the church and island of S. Giorgio Maggiore, which are beautiful from every point and at every hour, so happily do dome and white facade, red campanile and green roof, windowed houses and little white towers, compose. But then, in Venice everything composes: an artist has but to paint what he sees. From the Piazzetta's shore you look diagonally to the right to the Dogana and the vast Salute and all the masts in the Giudecca canal; diagonally to the left is the Lido with a mile of dancing water between us and it.
The shore of the Piazzetta, or more correctly the Molo, is of course the spot where the gondolas most do congregate, apparently inextricably wedged between the twisted trees of this marine forest, although when the time comes—that is, when the gondolier is at last secured—easily enough detached. For there is a bewildering rule which seems to prevent the gondolier who hails you from being your oarsman, and if you think that the gondolier whom you hail is the one who is going to row you, you are greatly mistaken. It is always another. The wise traveller in Venice having chanced upon a good gondolier takes his name and number and makes further arrangements with him. This being done, on arriving at the Molo he asks if his man is there, and the name—let us say Alessandro Grossi, No. 91 (for he is a capital old fellow, powerful and cheerful, with a useful supply of French)—is passed up and down like a bucket at a fire. If Alessandro chances to be there and available, all is well; but if not, to acquire a substitute even among so many obviously disengaged mariners, is no joke.
Old Grossi is getting on in years, although still powerful. A younger Herculean fellow whom I can recommend is Ferdinando, No. 88. Ferdinando is immense and untiring, with a stentorian voice in which to announce his approach around the corners of canals; and his acquaintanceship with every soul in Venice makes a voyage with him an amusing experience. And he often sings and is always good-humoured.
All gondoliers are not so. A gondolier with a grudge can be a most dismal companion, for he talks to himself. What he says, you cannot comprehend, for it is muttered and acutely foreign, but there is no doubt whatever that it is criticism detrimental to you, to some other equally objectionable person, or to the world at large.
The gondolier does not differ noticeably from any other man whose business it is to convey his fellow creatures from one spot to another. The continual practice of assisting richer people than oneself to do things that oneself never does except for a livelihood would seem to engender a sardonic cast of mind. Where the gondolier chiefly differs from, say, the London cabman, is in his gift of speech. Cabmen can be caustic, sceptical, critical, censorious, but they do occasionally stop for breath. There is no need for a gondolier ever to do so either by day or night; while when he is not talking he is accompanying every movement by a grunt.
It is this habit of talking and bickering which should make one very careful in choosing a lodging. Never let it be near a traghetto; for at traghetti there is talk incessant, day and night: argument, abuse, and raillery. The prevailing tone is that of men with a grievance. The only sound you never hear there is laughter.
The passion for bickering belongs to watermen, although loquacity is shared by the whole city. The right to the back answer is one which the Venetian cherishes as jealously, I should say, as any; so much so that the gondolier whom your generosity struck dumb would be an unhappy man in spite of his windfall.
The gondolier assimilates to the cabman also in his liking to be overpaid. The English and Americans have been overpaying him for so many years that to receive now an exact fare from foreigners fills him with dismay. From Venetians, who, however, do not much use gondolas except as ferry boats, he expects it; but not from us, especially if there is a lady on board, for she is always his ally (as he knows) when it comes to pay time. A cabman who sits on a box and whips his horse, or a chauffeur who turns a wheel, is that and nothing more; but a gondolier is a romantic figure, and a gondola is a romantic craft, and the poor fellow has had to do it all himself, and did you hear how he was panting? and do look at those dark eyes! And there you are! Writing, however, strictly for unattended male passengers, or for strong-minded ladies, let me say (having no illusions as to the gondolier) that every gondola has its tariff, in several languages, on board, and no direct trip, within the city, for one or two persons, need cost more than one franc and a half. If one knows this and makes the additional tip sufficient, one is always in the right and the gondolier knows it.
One of the prettiest sights that I remember in Venice was, one Sunday morning, a gondolier in his shirt sleeves, carefully dressed in his best, with a very long cigar and a very black moustache and a flashing gold ring, lolling back in his own gondola while his small son, aged about nine, was rowing him up the Grand Canal. Occasionally a word of praise or caution was uttered, but for the most part they went along silently, the father receiving more warmth from the consciousness of successful paternity than we from the sun itself.
Gondoliers can have pride: but there is no pride about a rampino, the old scaramouch who hooks the gondola at the steps. Since he too was once a gondolier this is odd. But pride and he are strangers now. His hat is ever in his hand for a copper, and the transference of your still burning cigar-end to his lips is one of the most natural actions in the world.
THE GRAND CANAL. I: FROM THE DOGANA TO THE PALAZZO REZZONICO, LOOKING TO THE LEFT
The river of Venice—Canal steamers—Motor boats—Venetian nobility to-day—The great architects—A desirable enactment—The custom house vane—The Seminario and Giorgione—S. Maria della Salute—Tintoretto's "Marriage in Cana"—The lost blue curtain—San Gregorio—The Palazzo Dario—Porphyry—The story of S. Vio—Delectable homes—Browning in Venice—S. Maria della Carita.
To me the Grand Canal is the river of Venice—its Thames, its Seine, its Arno. I think of it as "the river." The rest are canals. And yet as a matter of fact to the Venetians the rest are rivers—Rio this and Rio that—and this the canal.
During a stay in Venice of however short a time one is so often on the Grand Canal that a knowledge of its palaces should come early. For fifteen centimes one may travel its whole length in a steamboat, and back again for another fifteen, and there is no more interesting half-hour's voyage in the world. The guide books, as a rule, describe both banks from the same starting-point, which is usually the Molo. This seems to me to be a mistake, for two reasons. One is that even in a leisurely gondola "all'ora" one cannot keep pace with literature bearing on both sides at once, and the other is that since one enters Venice at the railway station it is interesting to begin forthwith to learn something of the city from that point and one ought not to be asked to read backwards to do this. In this book therefore the left bank, from the custom house to the railway station, is described first, and then the other side returning from the station to the Molo.
The Grand Canal has for long had its steamers, and when they were installed there was a desperate outcry, led by Ruskin. To-day a similar outcry is being made against motor-boats, with, I think, more reason, as I hope to show later. But the steamer is useful and practically unnoticeable except when it whistles. None the less it was an interesting experience in April of this year (1914) to be living on the Grand Canal during a steamer strike which lasted for several days. It gave one the quieter Venice of the past and incidentally turned the gondoliers into plutocrats.
But there is a great difference between the steamers and the motor-boat. The steamer does not leave the Grand Canal except to enter the lagoon; and therefore the injustice that it does to the gondolier is limited to depriving him of his Grand Canal fares. The motor-boat can supersede the gondola on the small canals too. It may be urged that the gondolier has only to become an engineer and his position will be as secure. That may be true; but we all know how insidious is the deteriorating influence of petrol on the human character. The gondolier even now is not always a model of courtesy and content; what will he be when the poison of machinery is in him?
But there are graver reasons why the motor-boat should be viewed by the city fathers with suspicion. One is purely aesthetic, yet not the less weighty for that, since the prosperity of Venice in her decay resides in her romantic beauty and associations. The symbol of these is the gondola and gondolier, indivisible, and the only conditions under which they can be preserved are quietude and leisure. The motor-boat, which is always in a hurry and which as it multiplies will multiply hooters and whistles, must necessarily destroy the last vestige of Venetian calm. A second reason is that a small motor-boat makes a bigger wash than a crowded Grand Canal steamer, and this wash, continually increasing as the number of boats increases, must weaken and undermine the foundations of the houses on each side of the canals through which they pass. The action of water is irresistible. No natural law is sterner than that which decrees that restless water shall prevail.
Enjoyment of voyages up and down the Grand Canal is immensely increased by some knowledge of architecture; but that subject is so vast that in such a hors d'oeuvre to the Venetian banquet as the present book nothing of value can be said. Let it not be forgotten that Ruskin gave years of his life to the study. The most I can do is to name the architects of the most famous of the palaces and draw the reader's attention to the frequency with which the lovely Ducal gallery pattern recurs, like a theme in a fugue, until one comes to think the symbol of the city not the winged lion but a row of Gothic curved and pointed arches surmounted by circles containing equilateral crosses. The greatest names in Venetian architecture are Polifilo, who wrote the Hypnerotomachia, the two Bons, Rizzo, Sansovino, the Lombardis, Scarpagnino, Leopardi, Palladio, Sammicheli, and Longhena.
In the following notes I have tried to mention the place of practically every rio and every calle so that the identification of the buildings may be the more simple. The names of the palaces usually given are those by which the Venetians know them; but many, if not more, have changed ownership more than once since those names were fixed.
Although for the most part the palaces of the Grand Canal have declined from their original status as the homes of the nobility and aristocracy and are now hotels, antiquity stores, offices, and tenements, it not seldom happens that the modern representative of the great family retains the top floor for an annual Venetian sojourn, living for the rest of the year in the country.
I wish it could be made compulsory for the posts before the palaces to be repainted every year.
And so begins the voyage. The white stone building which forms the thin end of the wedge dividing the Grand Canal from the Canale della Giudecca is the Dogana or Customs House, and the cape is called the Punta della Salute. The figure on the Dogana ball, which from certain points has almost as much lightness as Gian Bologna's famous Mercury, represents Fortune and turns with the wind. The next building (with a green and shady garden on the Giudecca side) is the Seminario Patriarcale, a great bare schoolhouse, in which a few pictures are preserved, and, downstairs, a collection of ancient sculpture. Among the pictures is a much dam-aged classical scene supposed to represent Apollo and Daphne in a romantic landscape. Giorgione's name is often associated with it; I know not with what accuracy, but Signor Paoli, who has written so well upon Venice, is convinced, and the figure of Apollo is certainly free and fair as from a master's hand. Another picture, a Madonna and Child with two companions, is called a Leonardo da Vinci; but Baedeker gives it to Marco d'Oggiano. There is also a Filippino Lippi which one likes to find in Venice, where the prevailing art is so different from his. One of the most charming things here is a little relief of the manger; as pretty a rendering as one could wish for. Downstairs is the tomb of the great Jacopo Sansovino.
And now rises the imposing church of S. Maria della Salute which, although younger than most of the Venetian churches, has taken the next place to S. Mark's as an ecclesiastical symbol of the city. To me it is a building attractive only when seen in its place as a Venetian detail; although it must always have the impressiveness of size and accumulation and the beauty that white stone in such an air as this can hardly escape. Seen from the Grand Canal or from a window opposite, it is pretentious and an interloper, particularly if the slender and distinguished Gothic windows of the apse of S. Gregorio are also visible; seen from any distant enough spot, its dome and towers fall with equal naturalness into the majestic Venetian pageant of full light, or the fairy Venetian mirage of the crepuscle.
The church was decreed in 1630 as a thankoffering to the Virgin for staying the plague of that year. Hence the name—S. Mary of Salvation. It was designed by Baldassarre Longhena, a Venetian architect who worked during the first half of the seventeenth century and whose masterpiece this is.
Within, the Salute is notable for possessing Tintoretto's "Marriage in Cana," one of the few pictures painted by him in which he allowed himself an interval (so to speak) of perfect calm. It is, as it was bound to be in his hands and no doubt was in reality, a busy scene. The guests are all animated; the servitors are bustling about; a number of spectators talk together at the back; a woman in the foreground holds out a vessel to the men opposite to show them the remarkable change which the water has undergone. But it is in the centre of his picture (which is reproduced on the opposite page) that the painter has achieved one of his pleasantest effects, for here is a row of pretty women sitting side by side at the banquetting table, with a soft light upon them, who make together one of the most charming of those rare oases of pure sweetness in all Tintoretto's work. The chief light is theirs and they shine most graciously in it.
Among other pictures are a S. Sebastian by Basaiti, with a good landscape; a glowing altar-piece by Titian, in his Giorgionesque manner, representing S. Mark and four saints; a "Descent of the Holy Ghost," by the same hand but under no such influence; and a spirited if rather theatrical "Nativity of the Virgin" by Lucia Giordano. In the outer sacristy the kneeling figure of Doge Agostino Barbarigo should be looked for.
The Salute in Guardi's day seems to have had the most entrancing light blue curtains at its main entrance, if we may take the artist as our authority. See No. 2098 in the National Gallery, and also No. 503 at the Wallace collection. But now only a tiny side door is opened.
A steamboat station, used almost wholly by visitors, is here, and then a canal, and then the fourteenth-century abbey of S. Gregorio, whose cloisters now form an antiquity store and whose severe and simple apse is such a rebuke to Longhena's Renaissance floridity. Next is a delightful little house with one of the old cup-chimneys, forming one of the most desirable residences in Venice. It has a glazed loggia looking down to the Riva. We next come to a brand new spacious building divided into apartments, then a tiny house, and then the rather squalid Palazzo Martinengo. The calle and traghetto of S. Gregorio, and two or three old palaces and the new building which now holds Salviati's glass business, follow. After the Rio del Formase is a common little house, and then the Palazzo Volkoff, once Eleonora Duse's Venetian home.
Next is the splendid fifteenth-century Palazzo Dario, to my eyes perhaps the most satisfying of all, with its rich colouring, leaning walls, ancient chimneys and porphyry decorations. Readers of Henri de Regnier's Venetian novel La Peur de l'Amour may like to know that much of it was written in this palace. We shall see porphyry all along the Canal on both sides, always enriching in its effect. This stone is a red or purple volcanic rock which comes from Egypt, on the west coast of the Red Sea. The Romans first detected its beauty and made great use of it to decorate their buildings.
Another rio, the Torreselle, some wine stores, and then the foundations of what was to have been the Palazzo Venier, which never was built. Instead there are walls and a very delectable garden—a riot of lovely wistaria in the spring—into which fortunate people are assisted from gondolas by superior men-servants. A dull house comes next; then a stoffe factory; and then the Mula Palace, with fine dark blue poles before it surmounted by a Doge's cap, and good Gothic windows. Again we find trade where once was aristocracy, for the next palace, which is now a glass-works' show-room, was once the home of Pietro Barbarigo, Patriarch of Venice.
The tiny church of S. Vio, now closed, which gives the name to the Campo and Rio opposite which we now are, has a pretty history attached to it. It seems that one of the most devoted worshippers in this minute temple was the little Contessa Tagliapietra, whose home was on the other side of the Grand Canal. Her one pleasure was to retire to this church and make her devotions: a habit which so exasperated her father that one day he issued a decree to the gondoliers forbidding them to ferry her across. On arriving at the traghetto and learning this decision, the girl calmly walked over the water, sustained by her purity and piety.
The next palace, at the corner, is the Palazzo Loredan where the widow of Don Carlos of Madrid now lives. The posts have Spanish colours and a magnificent man-servant in a scarlet waistcoat often suns himself on the steps. Next is the comfortable Balbi Valier, with a motor launch called "The Rose of Devon" moored to its posts, and a pleasant garden where the Palazzo Paradiso once stood; and then the great and splendid Contarini del Zaffo, or Manzoni, with its good ironwork and medallions and a charming loggia at the side. Robert Browning tried to buy this palace for his son. Indeed he thought he had bought it; but there was a hitch. He describes it in a letter as "the most beautiful house in Venice." The next, the Brandolin Rota, which adjoins it, was, as a hotel, under the name Albergo dell'Universo, Browning's first Venetian home. Later he moved to the Zattere and after that to the Palazzo Rezzonico, to which we are soon coming, where he died.
Next we reach the church, convent and Scuola of S. Maria della Carita, opposite the iron bridge, which under rearrangement and restoration now forms the Accademia, or Gallery of Fine Arts, famous throughout the world for its Titians, Tintorettos, Bellinis, and Carpaccios. The church, which dates from the fifteenth century, is a most beautiful brown brick building with delicate corbelling under the eaves. Once there was a campanile too, but it fell into the Grand Canal some hundred and seventy years ago, causing a tidal wave which flung gondolas clean out of the water. We shall return to the Accademia in later chapters: here it is enough to say that the lion on the top of the entrance wall is the most foolish in Venice, turned, as it has been, into a lady's hack.
The first house after the Accademia is negligible—newish and dull with an enclosed garden; the next is the Querini; the next the dull Mocenigo Gambara; and then we come to the solid Bloomsbury-blackened stone Palazzo Contarini degli Scrigni and its neighbours of the same ownership. Then the Rio S. Trovaso, with a pretty garden visible a little way up, and then a gay new little home, very attractive, with a strip of garden, and next it the fifteenth-century Loredan. A tiny calle, and then the low Dolfin. Then the Rio Malpaga and after it a very delectable new residence with a terrace. A calle and traghetto, with a wall shrine at the corner, come next, and two dull Contarini palaces, one of which is now an antiquity store, and then the Rio S. Barnaba and the majestic sombre Rezzonico with its posts of blue and faded pink.
This for long was the home of Robert Browning, and here, as a tablet on the side wall states, he died. "Browning, Browning," exclaim the gondoliers as they point to it; but what the word means to them I cannot say.
THE GRAND CANAL. II: BROWNING AND WAGNER
The Palazzo Rezzonico—Mr. and Mrs. Browning—Browning's Venetian routine—In praise of Goldoni—Browning's death—A funeral service—Love of Italy—The Giustiniani family—A last resource—Wagner in Venice—Tristan und Isolde—Plays and Music—The Austrians in power—The gondoliers' chorus—The Foscari Palace.
The Rezzonico palace and one of the Giustiniani palaces which are its neighbours have such interesting artistic associations that they demand a chapter to themselves.
Browning is more intimately associated with Florence and Asolo than with Venice; but he enjoyed his later Venetian days to the full. His first visit here in 1851, with his wife, was however marred by illness. Mrs. Browning loved the city, as her letters tell. "I have been," she wrote, "between heaven and earth since our arrival at Venice. The heaven of it is ineffable. Never had I touched the skirts of so celestial a place. The beauty of the architecture, the silver trails of water up between all that gorgeous colour and carving, the enchanting silence, the moonlight, the music, the gondolas—I mix it all up together, and maintain that nothing is like it, nothing equal to it, not a second Venice in the world."
Browning left Florence for ever after his wife's death, and to Venice he came again in 1878, with his sister, and thereafter for some years they returned regularly. Until 1881 their home was at the Brandolin Rota. After that they stayed with Mrs. Arthur Bronson, to whom he dedicated Asolando, his last book, and who has written a record of his habits in the city of the sea. She tells us that he delighted in walking and was a great frequenter of old curiosity shops. His especial triumph was to discover a calle so narrow that he could not put up an umbrella in it. Every morning he visited the Giardini Pubblici to feed certain of the animals; and on every disengaged afternoon he went over to the Lido, to walk there, or, as Byron had done, to ride. On being asked by his gondolier where he would like to be rowed, he always said, "Towards the Lido," and after his failure to acquire the Palazzo Manzoni he thought seriously for a while of buying an unfinished Lido villa which had been begun for Victor Emmanuel. Browning's desire was to see sunsets from it.
Mrs. Bronson tells us that the poet delighted in the seagulls, which in stormy weather come into the city waters. He used to wonder that no books referred to them. "They are more interesting," he said, "than the doves of St. Mark." Venice did not inspire the poet to much verse. There is of course that poignant little drama entitled "In a Gondola," but not much else, and for some reason the collected works omit the sonnet in honour of Goldoni which was written for the ceremonies attaching to the erection of the dramatist's statue near the Rialto. Mrs. Orr tells us that this sonnet, which had been promised for an album in praise of Goldoni, was forgotten until the messenger from the editor arrived for the copy. Browning wrote it while the boy waited. The day was November 27, 1883.
Goldoni—good, gay, sunniest of souls— Glassing half Venice in that verse of thine— What though it just reflect the shade and shine Of common life, nor render, as it rolls, Grandeur and gloom? Sufficient for thy shoals Was Carnival: Parini's depths enshrine Secrets unsuited to that opaline Surface of things which laughs along thy scrolls. There throng the people: how they come and go, Lisp the soft language, flaunt the bright garb,—see,— On Piazza, Calle, under Portico And over Bridge! Dear king of Comedy, Be honoured! Thou that did'st love Venice so, Venice, and we who love her, all love thee.
The Rezzonico is the house most intimately associated with Browning in the public mind, although most of his Venetian life was spent elsewhere. It was here, on his last visit to his son, that the poet died. He had not been very well for some time, but he insisted on taking his daily walk on the Lido even although it was foggy. The fog struck in—it was November—and the poet gradually grew weaker until on December 12, 1889, the end came. At first he had lain in the left-hand corner room on the ground floor; he died in the corresponding room on the top floor, where there was more light.
Browning was buried in Westminster Abbey, but a funeral service was held first in Venice. In his son's words, "a public funeral was offered by the Municipality, which in a modified form was gratefully accepted. A private service, conducted by the British Chaplain, was held in one of the halls of the Rezzonico. It was attended by the Syndic of Venice and the chief City authorities, as well as by officers of the Army and Navy. Municipal Guards lined the entrance of the Palace, and a Guard of Honour, consisting of City firemen in full dress, stood flanking the coffin during the service, which was attended by friends and many residents. The subsequent passage to the mortuary island of San Michele was organized by the City, and when the service had been performed the coffin was carried by firemen to the massive and highly decorated funeral barge, on which it was guarded during the transit by four 'Uscieri' in gala dress, two sergeants of the Municipal Guard, and two firemen bearing torches. The remainder of these followed in their boats. The funeral barge was slowly towed by a steam launch of the Royal Navy. The chief officers of the Municipality, the family, and many others in a crowd of gondolas, completed the procession. San Michele was reached as the sun was setting, when the firemen again received their burden and bore it to the principal mortuary chapel."
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 563
|
<!DOCTYPE html>
{%- include snippets/get-lang.html -%}
<html lang="{{ __return }}">
<head>
{%- include common-head.html -%}
{%- include head-icons-rel.html -%}
{%- assign __path = '/assets/css/blog.css' -%}
{%- include snippets/prepend-baseurl.html -%}
<link rel="stylesheet" href="{{ __return }}">
<script src="https://code.jquery.com/jquery-3.3.1.min.js"></script>
</head>
<body>
<div class="m-page-stage js-page-stage">{{ content }}</div>
{%- include components/footer.html -%}
<script>
$(function() {
// display coding language
var $articleContent = $('.m-post, .m-page').find('.m-article-content'), $this;
$articleContent.children('.highlight').each(function() {
$this = $(this);
$this.attr('data-lang', $this.find('code').attr('data-lang'));
});
$articleContent.children('h1, h2, h3, h4, h5, h6').each(function() {
$this = $(this);
$this.append($('<a class="anchor" aria-hidden="true"></a>').text('#'));
});
$articleContent.on('click', '.anchor', function(e) {
var targetId = $(this).parent().attr('id');
scrollAnimateTo($('#' + targetId).offset().top, 400, function() {
console.log(666)
window.history.replaceState(null, '', window.location.href.split('#')[0] + '#' + targetId);
});
});
function scrollAnimateTo(destination, duration, callback) {
var $body = $('html, body'), bodyScrollTop = $body.scrollTop();
if(bodyScrollTop === destination) { return; }
$body.animate({ scrollTop: destination }, duration, callback);
}
});
</script>
{%- include utils/mathjax.html -%}
{%- include utils/mermaid.html -%}
{%- include utils/chart.html -%}
{%- if jekyll.environment != "development" -%}
{%- include utils/google-analytics.html -%}
{%- endif -%}
</body>
</html>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,055
|
Looking for a reason to give roses to your love ones?
Sometimes, there should not be any reason attached. Giving flowers to your better half, your mother or your love ones could be a random act of love just to show someone that you care. To show how much you care, you can head to the Roses Only for roses that are like no other in the market.
Roses Only is an online florist that retails mostly roses for all occasion.
Roses only comes in a box that is almost 3 feet tall. The reason - The roses provided are long stem roses that are not only unique but impressive as a show of love. Each roses are about 60-80cm tall!
There are 12 different colours to choose from. From evergreen colours like Freedom Red to specialty hues like Iguana Dark Orange, you would be able to choose one that fits the mood.
Roses starts from under $100 for a dozen roses to 100 roses for THAT special occasion.
The roses can last for a minimum of 5-7 days with proper care.
Interested to get one for your love ones?
TWD received a bouquet of roses for review purpose only.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 6,620
|
Константи́н Инноке́нтьевич Помера́нцев (, Красноярск — 21 мая 1945, Курск) — русский и советский живописец, скульптор, архитектор.
Биография
До революции
Померанцев происходил из семьи чиновника. В 1893—1897 годах он проходил обучение в Красноярской мужской гимназии. В 1900 семья Померанцева переехала в Верхоленский уезд Иркутской губернии, но после смерти отца, наступившей в 1906 году, вся многодетная семья переехала в Иркутск, где Померанцев, начиная с 1902 года, обучался в Иркутском горном училище, которое закончил в 1907 году. В 1906—1907 годах он проходил обучение в классах рисования и живописи А. Ф. Лытнева, а в 1908—1910 годах Померанцев обучался в Москве живописи в художественном училище Ф. И. Рерберга и в то же самое время он брал уроки скульптуры у А. С. Голубкиной.
В 1910 году Померанцев вернулся в Иркутск и поступил на службу в Контроле Забайкальской железной дороги в должности помощника контролёра, в которой он оставался вплоть до 1920 года. Померанцев стал создателем (в 1910 году) и руководителем Иркутского общества художников, которое он возглавлял до 1923 года (когда общество было реорганизовано).
Уже в первый свой период проживания в Иркутске Померанцев стал одним из заметных деятелей художественной жизни города (принимал участие в Первой передвижной художественной выставки для Сибири в 1903 году), а по возвращении из Москвы он участвовал в художественных выставках в музее Восточно-Сибирского отдела Императорского русского географического общества (во 2-й (1910), 3-й (1911), 4-й (1912), 5-й (1913) и 6-й (1914)), а также в 1-й (1915), 2-й (1916) и 3-й (1917) выставках Иркутского общества художников, выставке картин сибирских художников в Иркутске в 1912 году, выставке Приамурского края, посвящённой 300-летию царствования дома Романовых, состоявшейся в Хабаровске в 1913 году, где был удостоен похвального листа, а также принял участие в VI и VIII периодических выставках Томского общества любителей художеств в декабре 1913 года — январе 1914 года и в декабре 1915 года — январе 1916 года, в благотворительной картинной выставке в Иркутске в 1915 году, Первой сибирской передвижной выставке живописи и скульптуры (1916 год) и Второй художественной выставке (1917 год) в Красноярске, художественной выставке, устроенной Томскими рисовальными классами в 1917 году и в ряде других.
В 1914—1916 годах Померанцев иллюстрировал выпускавшиеся в Иркутске издателем М. Е. Стожем железнодорожные «Спутники», а также издания книгоиздательства «Ирисы» и некоторые иркутские журналы («Иркутская незабудка», «Маленький железнодорожник» и, вероятно, другие).
Послереволюционный период
В 1920—1925 работал сначала 2-м, затем 1-м хранителем художественно-исторического отделения (картинной галереи) областного музея. Преподавал в художественной студии при Политотделе 5-й Армии (1920—1922). В 1920—1923 годах возглавлял Иркутское общество художников, пока в 1923 году оно не было ликвидировано в результате реорганизации творческих объединений в единый союз. В январе 1920, после победы над Белой Армией, Померанцев возглавил коллектив художников по оформлению города к торжественной встрече частей 5-й Армии, которая состоялась 10 марта 1920.
21 февраля 1920 иркутским революционным комитетом была национализирована картинная галерея городского головы Иркутска В. П. Сукачёва. Померанцев произвёл приём и опись всех произведений, и до своего отъезда в Монголию в 1926 году был главным хранителем этой галереи.
Померанцев принимал участие в выставках: весенней и осенней выставках Иркутского общества художников (ИОХ) в 1919 году, весенних выставках ИОХ в 1920 и 1922 годах, краеведческой выставке в 1925 году, групповой выставке иркутских художников в 1925 году, Первой Всесибирской художественной выставке в 1927 году. В 1923 Померанцев устроил в Иркутске персональную выставку.
В 1925 Померанцев оформил в китайском стиле фасад и интерьер магазина «Чаеуправления» (на углу улиц Урицкого и Карла Маркса, оформление было уничтожено в 1967 году). В архитектурном оформлении этого сооружения большое участие принимала супруга Померанцева В. И. Шаршун.
Монгольский период
В 1926 Померанцев был приглашён монгольским правительством в Улан-Батор для художественного оформления монгольской столицы, в частности для создания памятника Сухэ–Батору, который был установлен на центральной площади Улан-Батора в 1930 году, а также создания музея Монгольской революции.
В это же время в Монголии находился Н. Рерих с которым у Померанцева сложились дружеские отношения. На главной площади Улан-Батора строилось здание так называемого «Народного дома» (Нардома), в нём проходили сессии парламента Монголии (хуралы) и ставили первые спектакли. Монгольским правительством было сделано предложило Н. К. Рериху расписать первый занавес для него, однако тот порекомендовал на эту работу Померанцева, картины которого он ценил. Померанцев изобразил на занавесе уличного музыканта Ёндон-хурчи с национальным инструментом моринхуром.
Померанцев был художником-оформителем спектаклей театральной студии Народного дома и спектаклей образованного в 1931 Государственного центрального народного театра Монголии. Он также преподавал рисование в школах Улан-Батора, организовал и вёл художественную студию, его учениками были известные в будущем монгольские художники Л. Гава, О. Цэвэгжав и другие. Вместе со своей женой, художницей и скульптором Верой Ивановной Шаршун, устроил несколько выставок своих работ (1928, 1934, 1937, 1938, 1939 и др.).
Померанцев много путешествовал по Монголии, делал зарисовки, принимал деятельное участие в просветительской работе среди населения, устраивал выставки. Он изучил монгольский язык и читал лекции по искусству. Померанцев принимал участие в археологических экспедициях. В Монголии им был собран богатый материал по истории страны, а на основании археологических данных написаны картины «Угеде» («Сын Чингисхана»), «В Каракоруме» и «Каменная черепаха» («Уйгуры») (1937). Быту и старинным обычаям монгольского народа посвящены его картины «Перерожденец» и «Там, где хоронят».
В историко-революционной тематике были созданы картины «Сухэ-Батор перед партизанами» и «Изгнание интервентов». В тематике общественно-политической были написаны картины «Украшение к Октябрю», «Кооператив в уезде», «Строительство комбината», «Парад», «Школьники». В столице Монголии Померанцев помимо скульптурного памятника Сухэ-Батору создал большой скульптурный барельеф для фасада здания Дворца культуры имени Ленина в Улан-Баторе. Произведения Померанцева и его жены В. И. Шаршун, созданные в период работы в Монголии, были представлены на выставке, которая состоялась сначала в Монголии (1939), а затем в Иркутске (1940).
В общей сложности Померанцев прожил в Монголии 13 лет. Более 350 картин, этюдов, рисунков, набросков Померанцева хранятся в Монгольском театральном музее (в городе Улан-Батор).
В Улан-Баторе родилась дочь Померанцева и В. И. Шаршун Зоя (1934 год).
1939—1945 годы
Осенью 1939 вернулся в Иркутск, в том же году принял участие в областной выставке иркутских художников, посвященной двадцатилетию освобождения Сибири от войск Колчака. В феврале — марте 1940 года в художественном музее состоялась выставка произведений Померанцева (совместно с его женой В. И. Шаршун), созданных за период их работы в Монголии.
В 1940—1944 годах Померанцев жил в селе Палех Ивановской области, где он был директором Палехского музея и одновременно преподавателем Палехского художественного училища. В этот период Померанцев писал статьи, относящиеся к тематической области культуры и искусства, в местной газете «Трибуна Палеха». Померанцев с 1941 года был членом Ивановского отделения Союза советских художников, участвовал в областных выставках в городе Иваново.
В 1944 году Померанцев переехал в Курск, где участвовал в курской областной выставке художников (в 1945 году).
Галерея
Личная жизнь
Померанцев был женат на художнике и скульпторе Вере Ивановне Шаршун (1899, Бугуруслан — 1978, Москва), сестре Сергея Шаршуна — русского писателя и художника.
Имел дочь Зою (в замужестве Виницкую; 1934, Улан-Батор — 1992, Москва).
Примечания
Ссылки
Померанцев, Константин Иннокентьевич ИркиПедия
Померанцев Константин Иннокентьевич Культура Приангарья.
Померанцев Константин Иннокентьевич История и культура Сибири
Померанцев Константин Иннокентьевич АртРу.инфо
Померанцев Константин Иннокентьевич Бабр.ру
Литература
Померанцев Константин Иннокентьевич (1884-1945) / Т. Г. Ларева // Художники Иркутска. - Иркутск : Восточно-Сибирское книжное изд-во, 1994. - С. 23-27 : цв.ил. ББК 85.143(2)
Художник Константин Померанцев : Письма. Документы. Воспоминания / Архитектур.-этнограф. музей "Тальцы" ; Сост. Ю. П. Лыхин; Ред. И. Р. Хенох. - Иркутск : Гос. учреждение Архитектур.-этнограф. музей "Тальцы", 2001 (Иркутск : Изд-во ИГЭА, 2001). - 147 с. : фот., ил. ; 21 см. - Библиогр. : с. 145 (21 назв.). - 300 экз.. - ISBN 5-94318-003-6
Память, звучащая в красках / Ю. Башинова // Вост.-Сиб. правда : обществ.-полит. газ. Иркут. обл. - 2014. - № 38(16-23 сент.). - С. 10 : фот. ББК 85.14
Художники Иркутска
Архитекторы Иркутска
Художники Монголии
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 1,441
|
\section{Introduction}
\label{sec:intro}
The JPEG format is the de facto compression standard for digital images in
a wide range of fields from medical imaging to personal digital cameras. By
our own observation, 463 out of the 500 most popular
websites~\cite{top500sites} use JPEG images (including Google, Facebook,
Youtube and Baidu). With the emergence of social networks and innovations in
computational photography, billions of JPEG images are shared and viewed on a
daily basis: Facebook reported in 2010 already to store over 65~billion
photos~\cite{Beaver10findinga}, and the Instagram photo-sharing service claims
to have 45~million daily uploads~\cite{InstagramPress}. Similar numbers can be
assumed for other photo hosting and sharing services such as Flickr and
Photobucket.
Desktops, tablets and smartphones constitute the vast majority of hardware
platforms used for viewing JPEG images. Although these platforms are nowadays
equipped with a CPU and GPU, to the best of our knowledge no approach is
available yet that is capable of incorporating both the CPU and
the GPU for JPEG decoding.
Libjpeg~\cite{libjpeg} is a sequential JPEG reference implementation
by the Independent JPEG Group. To accelerate image processing, the
libjpeg-turbo~\cite{libjpegturbo} re-implementation applies SIMD instructions
with a single thread of execution on
x86 and ARM platforms. We have observed that the SIMD-version of libjpeg-turbo
decodes
an image twice as fast as the sequential version on an Intel~i7. Libjpeg-turbo is
widely used, e.g., with the Google Chrome and Firefox web-browsers,
WebKit~\cite{webkit}, and the Ubuntu, Fedora and openSUSE Linux distributions.
Neither libjpeg nor libjpeg-turbo are capable of utilizing a GPU.
JPEG decoding is computationally expensive, consisting of Huffman
decompression, dequantization, IDCT,
image upsampling and YCbCr to RGB color space conversion. Among all stages, Huffman
decompression is strictly sequential, because code-words have variable
lengths and the start of a codeword in the encoded bitstream is only
known once the previous codeword has been decoded.
A sub-class of Huffman codes that provide the so-called
self-synchronization
property~\cite{huffman1988sync} are suitable for decoding multiple
chunks of the encoded bitstream in parallel, as
proposed in~\cite{Klein03parallelhuffman}. However, the JPEG standard
does not enforce the self-synchronization property~\cite{Wallace91thejpeg}.
In our implementation, Huffman decoding
is therefore executed sequentially on the CPU. The remaining stages have repetitive
computations and low data dependencies, which makes them suitable to
exploit data, task and pipeline-parallelism.
A desktop GPU has several hundreds of scalar processors, offering more
parallelism than what is provided by nowaday's SIMD CPU instruction
set architectures. GPUs offer a higher memory bandwidth than
CPUs.
However, a GPU core lacks complex control units and operates at a much lower
clock frequency. The PCI bus that connects the GPU to the CPU represents a
bandwidth-bottleneck that incurs significant overhead to computations on the
GPU. JPEG decoding on a GPU is thus not necessity faster than the SIMD-version
of libjpeg-turbo on a CPU. Nevertheless, utilizing both CPU and GPU has the
potential to achieve the highest overall performance, regardless of the
computational power of the CPU and GPU.
Consequently, this paper makes the following contributions:
\begin{itemize}
\item
we propose a performance model based on off-line profiling that characterizes
all JPEG decoding steps for a given CPU-GPU combination
from multivariate polynomial regression.
We identified image entropy and the image dimensions as the only
required parameters for our performance model,
\item
we propose a dynamic partitioning scheme that automatically
distributes the workload across CPU and GPU at run-time according to
our performance model,
\item
we optimize JPEG decoding on GPUs by employing data vectorization,
intermediate local memory and coalesced memory accesses,
\item
we boost
parallelism by utilizing a pipelined execution model that overlaps sequential
Huffman decoding with GPU computations,
\item
we report on the software
engineering aspects of refactoring the libjpeg-turbo legacy code for
heterogeneous multicores, and
\item
we present experimental results on three
representative high- and mid-end CPU-GPU architectures for the applicability
and efficiency of our approach. We achieve speedups up to 4.2x over the
SIMD-version of libjpeg-turbo, and up to 8.5x over its sequential code.
We achieve up to 95\% of the theoretically attainable
speedup, with 88\% on average.
\end{itemize}
The remainder of this paper is organized as follows:
Section~\ref{sec:jpegdecoding} presents background information on JPEG
decoding; libjpeg-turbo re-engineering is discussed in
Section~\ref{sec:reengineering}. Our OpenCL JPEG decoding kernels for GPUs
are presented in Section~\ref{sec:paralleljpeg}.
Section~\ref{sec:heterogjpeg} describes the performance model and dynamic
partitioning scheme for a system consisting of a CPU and a GPU.
Section~\ref{sec:experiment} contains the experiential
results. We discuss the related work in Section~\ref{sec:related}
and draw our conclusions in Section~\ref{sec:conclusions}.
\section{Background: JPEG Decoding}
\label{sec:jpegdecoding}
Figure~\ref{fig:decoderpath} describes the decoding steps to produce an
uncompressed bitmap from a JPEG image. A JPEG file is structured
as a sequence of segments, including image dimensions, component subsampling,
Huffman and quantization tables and entropy-coded data. Entropy-coded data
is the largest part of a JPEG file, and thus, has the highest contribution to
the file size. Color in JPEG-encoded images is represented by luminance (Y),
blue chrominance (Cb) and red chrominance (Cr) component values. Because the
human eye is
more sensitive to changes in luminance than changes in chrominance, the
spatial resolution of chrominance components are commonly compressed. This
process is called downsampling. In 4:2:2 subsampling, the Y component is sampled
at each pixel while Cb and Cr components are sampled every two pixels in
horizontal direction. 4:4:4 sampling has the same
sample rate across all components and downsampling is not
required~\cite{poynton2002chroma}.
\tikzstyle{djpegrect}=[rectangle, draw=black, rounded corners, fill=white, drop shadow,
text centered, anchor=north, text=black, text width=1.9cm,minimum height=1cm]
\tikzstyle{djpegarrow}=[line width=2pt,draw, -latex]
\begin{figure}[htb]
\begin{center}
{\scalefont{0.9}
\begin{tikzpicture}[node distance=0.22cm]
\node (Huffman) [djpegrect, rectangle]
{Entropy Decoding (Huffman)};
\node (AuxNode01) [text width=0.22cm, right=of Huffman] {};
\node (Dequantization) [djpegrect, rectangle , right=of AuxNode01]
{De-quantization};
\node (AuxNode02) [text width=0.22cm, right=of Dequantization] {};
\node (IDCT) [djpegrect, rectangle, right=of AuxNode02]
{Inverse DCT};
\node (AuxNode03) [text width=0.22cm, below=of AuxNode02] {};
\node (Upsampling) [djpegrect, rectangle, below=of AuxNode03]
{Upsampling};
\node (AuxNode04) [text width=0.22cm, left=of Upsampling] {};
\node (Color) [djpegrect, rectangle, left=of AuxNode04]
{Color Conversion};
\draw[djpegarrow] (Huffman.east) -- (Dequantization.west);
\draw[djpegarrow] (Dequantization.east) -- (IDCT.west);
\draw[djpegarrow] (IDCT.south) |- (Upsampling.east);
\draw[djpegarrow] (Upsampling.west) -- (Color.east);
\end{tikzpicture}
}
\caption{JPEG decoder path.}
\label{fig:decoderpath}
\end{center}
\end{figure}
An image is divided into blocks of 8x8 pixels. JPEG decompression operates in
units of minimum coded units (MCUs), which are minimum sets of blocks from each color
component. The MCU size for 4:4:4 subsampling is 8x8 pixels in libjpeg-turbo.
In 4:2:2 subsampling, one chrominance block is upsampled to two blocks
horizontally. Thus, an MCU has a size of 16x8 pixels. The decoder first
decodes entropy data, then de-quantizes it according to an image-specific
quantization table. IDCT transforms MCUs from the frequency domain back to the
spatial domain. The libjpeg and libjpeg-turbo libraries apply a series of 1D
IDCTs based on the AAN
algorithm~\cite{yukihiro1988fast,JPEGcompressionStandard}. If the image
subsampling is not 4:4:4, it must be upsampled to restore the spatial
resolution of chrominance components to the original size. Color conversion
converts the Y, Cr and Cb samples of each pixel to the RGB color space. Apart
from Huffman decoding, all JPEG decoding steps contain few data dependencies and
a large amount of data-parallelism. In fact, libjpeg-turbo utilizes SIMD instructions
for all stages except Huffman decoding.
\section{Re-engineering the Libjpeg-turbo Software Architecture for Heterogeneous
Multicores}
\label{sec:reengineering}
Libjpeg-turbo has been designed with a consideration of memory
resources. Both encoder and decoder operate in units of MCU rows.
The software architecture of libjpeg-turbo is illustrated in Figure~\ref{fig:arch}.
The decoder uses a 2-tier controller and buffer hierarchy to control
decoding and storage of a single MCU row in various stages.
\tikzstyle{turborect}=[rectangle, draw=black, rounded corners, fill=white, drop shadow,
text centered, anchor=north, text=black, text width=1.5cm,minimum height=0.7cm]
\tikzstyle{turbocircle}=[circle, draw=black, fill=white, drop shadow,
text centered, anchor=north, text=black,minimum size=0.9cm]
\tikzstyle{line}=[draw]
\begin{figure}[htb]
\begin{center}
{\scalefont{0.8}
\begin{tikzpicture}[node distance=0.22cm]
\node (Main) [turborect, rectangle]
{Main Control};
\node (B1) [turbocircle, right=of Main]{Buffer};
\node (AuxNode01) [text width=2cm, right=of B1] {};
\node (Coefficient) [turborect, rectangle , above=0.6cm of AuxNode01]
{Coefficient Control};
\node (B2) [turbocircle, right=of Coefficient]{Buffer};
\node (AuxNode02) [text width=0.22cm, right=of B2] {};
\node (Postprocessing) [turborect, rectangle, below=0.6cm of AuxNode01]
{Post-processing Control};
\node (B3) [turbocircle, right=of Postprocessing]{Buffer};
\node (AuxNode03) [text width=0.22cm, right=of B3] {};
\node (AuxNode04) [text width=1cm, right=of AuxNode02] {};
\node (IDCT) [turborect, rectangle, below=0.1cm of AuxNode04,text width=1.5cm]
{Dequantize and inverse DCT};
\node (Huffman) [turborect, rectangle, above=of IDCT,text width=1.5cm]
{Entropy decoding};
\node (AuxNode05) [text width=1cm, right=of AuxNode03] {};
\node (Upsampling) [turborect, rectangle, above=0.1cm of AuxNode05,text width=1.5cm]
{Upsampling};
\node (Color) [turborect, rectangle, below=of Upsampling,text width=1.5cm]
{Color space Conversion};
\node (Cquantization) [turborect, rectangle, below=of Color,text width=1.5cm]
{Color quantization};
\node (Cprecision) [turborect, rectangle, below=of Cquantization,text width=1.5cm]
{Color precision reduction};
\path [line] (Main.east) -- (B1.west);
\path [line] (B1.east) -- (AuxNode01.west);
\path [line] (AuxNode01.west) |- (Coefficient.west) ;
\path [line] (AuxNode01.west) |- (Postprocessing.west) ;
\path [line] (Coefficient.east) -- (B2.west);
\path [line] (Postprocessing.east) -- (B3.west);
\path [line] (B2.east) -- (AuxNode02.west);
\path [line] (AuxNode02.west) |- (Huffman.west) ;
\path [line] (AuxNode02.west) |- (IDCT.west) ;
\path [line] (B3.east) -- (AuxNode03.west);
\path [line] (AuxNode03.west) |- (Upsampling.west) ;
\path [line] (AuxNode03.west) |- (Color.west) ;
\path [line] (AuxNode03.west) |- (Cquantization.west) ;
\path [line] (AuxNode03.west) |- (Cprecision.west) ;
\end{tikzpicture}
}
\caption{Libjpeg-turbo software architecture.}
\label{fig:arch}
\end{center}
\end{figure}
We identified two shortcomings which hamper parallelizing the library for
heterogeneous multicores. First, because decoding is done in units of MCU
rows, additional, unnecessary dependencies between subsequent MCU rows are
introduced. These dependencies limit the possible achievable parallelism in
three ways:
\begin{enumerate}
\item
A single MCU row of an image may not contain enough data to utilize a GPU. Most of
the computationally intensive operations, i.e., IDCT, upsampling and color
conversion, are data-parallel tasks where more data means more parallelism.
Equally worse, pipeline-parallelism between decoding steps
is impossible because of those dependencies.
\item Transferring row after row of image data from CPU to GPU is inefficient,
because initiating a transfer induces constant overhead. Transferring a large
amount of data in one transfer is thus more efficient than using several
smaller transfers.
\item
For each MCU row, a kernel invocation on the GPU is required.
\end{enumerate}
Second, for modularity reasons, the libjpeg-turbo library has been designed
in two major parts: coefficient control and postprocessing control.
A 2-tier buffer hierarchy abstracts away the actual decoding work,
and function pointers are used to encapsulate decoding functionality (e.g.,
integer vs.~float IDCT). The segmented software architecture makes it very
hard to re-use JPEG decoder components, because the buffer hierarchy
permeates all components.
We re-engineered the libjpeg-turbo library under two objectives: (1)~to be
minimally invasive to the legacy code, and (2)~to support massively parallel
architectures. To prevent expensive CPU-GPU data transfers, we introduced an
additional input and output buffer below the existing buffer hierarchy. These
buffers are large enough to keep an image as a whole in memory. (Note that
this fits in naturally with JPEG decoding, e.g., a web-browser will initiate
decoding once the whole image or a large part of it is available in main
memory.) The new whole-image buffers allowed us to transfer sufficiently large
chunks of the image between CPU and GPU, while providing the legacy-code on the
CPU with its ``accustomed'' row-by-row access, thereby keeping the changes to
the existing library code to a minimum. The kernel codes for the GPU were
implemented in OpenCL. The existing library code served as the starting point
for the GPU code, with all GPU-specific optimizations explained in
Section~\ref{sec:paralleljpeg}.
\section{JPEG Decoding on the GPU}
\label{sec:paralleljpeg}
After entropy (Huffman) decoding, the CPU transfers a buffer of decoded data to
the GPU. The IDCT, upsampling and color conversion kernels are invoked
subsequently. Our chosen buffer layout has Y blocks followed by Cb blocks
followed by Cr blocks. The upsampling kernel does not have to read the
Y-space. This buffer layout avoids interleaving block access, and thus,
improves coalesced memory access. At the end of color conversion, the output
image in RGB color is transferred to a designated memory location of the
whole-image output buffer (see Section~\ref{sec:reengineering}) on the CPU.
\subsection{Inverse Discrete Cosine Transformation (IDCT)}
\label{sub:IDCT}
The entropy-decoded data in the frequency domain is transformed back
to the spatial domain using a 2D IDCT.
We implemented the 2D IDCT algorithm by applying a 1D IDCT to eight columns of a block (column pass) and then to eight rows of the result (row pass), as shown in Equation~\eqref{columnpass} and Equation~\eqref{rowpass} respectively.
\begin{equation}
\label{columnpass}
f(u,y) = \sum_{v=0}^{N-1}C_vF(u,v)\cdot\cos\left(\frac{(2y+1)v\pi}{2N}\right),
\end{equation}
\begin{equation}
\label{rowpass}
f(x,y) = \sum_{u=0}^{N-1}C_uf(u,y)\cdot\cos\left(\frac{(2x+1)u\pi}{2N}\right),
\end{equation}
where
\begin{align*}
&0\le x,y \le N-1:\thinspace \text{spatial coordinates},\\
&0\le u,v \le N-1:\thinspace \text{frequency coordinates},\\
&C_u, C_v=\frac{1}{\sqrt{2}} \text{ for } u,v = 0, \text{ otherwise } 1.
\end{align*}
We employ eight OpenCL work-items per block. The input data is de-quantized
after being loaded from global memory. Each work-item performs the column pass
followed by the row pass. A work-item stores an eight-pixel column directly to
its registers such that no communication is required among work-items. The
intermediate results from the column pass are shared among work-items within a
group to process the row pass. Thus, local memory is the suitable choice.
Each work-item holds eight elements of 8-bit color representation at the end of
the row pass. Copying eight times would generate an excessive overhead. Hence, we
vectorize the elements to reduce global memory access requests.
Instructions are issued per group of work-items called a warp in NVIDIA's
terminology. The warp size is typically 32. Therefore, a work-group performs
IDCT on a multiple of four blocks to ensure that the number of work-items per
group is a multiple of 32. The optimal work-group size is hardware-specific
and is determined during profiling (see Section~\ref{sec:heterogjpeg}).
\vskip+5mm
\subsection{Upsampling}
The chrominance color space with 4:2:2 subsampling is downsampled to half of the
luminance space during JPEG encoding. The sample rates of these color spaces
must be upsampled to the original size. Algorithm~\ref{algo:UpSampling}
describes an upsampling process that takes an 8-pixel row as an input to
generate a 16-pixel row.
\begin{algorithm}[t]
\SetKwInOut{Input}{Input}
\SetKwInOut{Output}{Output}
\SetKwData{STATE}{state}
\SetKw{OUT}{Out}
Out[0] = In[0] \\
Out[1] = (In[0] * 3 + In[1] + 2) / 4 \\
Out[2] = (In[1] * 3 + In[0] + 1) / 4 \\
Out[3] = (In[1] * 3 + In[2] + 2) / 4 \\
Out[4] = (In[2] * 3 + In[1] + 1) / 4 \\
Out[5] = (In[2] * 3 + In[3] + 2) / 4 \\
Out[6] = (In[3] * 3 + In[2] + 1) / 4 \\
Out[7] = (In[3] * 3 + In[4] + 2) / 4 \\
Out[8] = (In[4] * 3 + In[3] + 1) / 4 \\
Out[9] = (In[4] * 3 + In[5] + 2) / 4 \\
Out[10] = (In[5] * 3 + In[4] + 1) / 4 \\
Out[11] = (In[5] * 3 + In[6] + 2) / 4 \\
Out[12] = (In[6] * 3 + In[5] + 1) / 4 \\
Out[13] = (In[6] * 3 + In[7] + 2) / 4 \\
Out[14] = (In[7] * 3 + In[6] + 1) / 4 \\
Out[15] = In[7]
\caption{Upsampling for 4:2:2 subsampling.}
\label{algo:UpSampling}
\end{algorithm}
We utilize 16 OpenCL work-items to perform upsampling on one block. Two
work-items process one row of the block. The work-item with the even ID reads
In[0] to In[4] to produce an eight-pixel row from Out[0] to Out[7], and the
work-item with the odd ID reads In[4] to In[7] to produce the successive
eight-pixel row Out[8] to Out[15].
The output equations have fixed patterns for odd indices and even indices.
It should be noted that all but the end pixels depend on the neighbouring
pixels of a block. The computations of Out[0] and Out[8] happen concurrently,
and the same situation occurs with Out[7] and Out[15]. The computational
pattern of the end pixels is different from the other pixels. Consequently,
an if-statement is required to determine the correct equation for a specific
work-item. An if-statement causes branch divergence if less than half of a
warp take the same branch. We chose the work-group size such that 16
work-items take the same branch. This access pattern was designed to favour a
merged upsampling-color conversion kernel, which is explained in
Section~\ref{KernelMerging}.
\vskip+5mm
\subsection{Color Conversion}
The final stage of the JPEG decoder converts the YCbCr color space to the RGB color
space according to Algorithm~\ref{algo:ColorSpaceConversion}. A work-item
accesses global memory three times for its Y, Cb and Cr values to calculate R, G and
B values for one pixel. The computations for each pixel are independent
of other pixels.
\begin{algorithm}[th]
\SetKwInOut{Input}{Input}
\SetKwInOut{Output}{Output}
\SetKwData{STATE}{state}
\SetKw{OUT}{Out}
\Input{Pixel information in YCbCr color space}
\Output{Pixel information in RGB color space}
R = Y + 1.402 (Cr - 128)\\
G = Y - 0.34414 (Cb - 128) - 0.71414 (Cr - 128)\\
B = Y + 1.772 (Cb - 128)
\caption{Cb/Cr to RGB color space conversion.}
\label{algo:ColorSpaceConversion}
\end{algorithm}
The buffers for IDCT and upsampling are arranged as a sequence of blocks, shown
in Figure~\ref{fig:bufferindexing:a}. However, the output buffer of color
conversion is arranged as a sequence of pixels starting from the top-left pixel
of the image then traverses row-wise to the bottom-right pixel of the image as
shown in Figure~\ref{fig:bufferindexing:b}. We devised an indexing
function that calculates the index of the next pixel in vertical
direction to be one image-width apart.
\begin{figure}[htb]
\begin{center}
\subfigure[Block-based pattern]
{
\begin{tikzpicture}[scale=0.6,every node/.style={scale=0.6}]
\foreach \i in {0,...,15} {
\draw [very thin,black!20,text=black] (\i/2,0) -- (\i/2,4) node [above] at (\i/2+0.25,4) {$\i$};}
\foreach \i in {0,...,7} {
\draw [very thin,black!20,text=black] (0,\i/2) -- (8,\i/2) node [left] at (0,4-\i/2-0.25) {$\i$};}
\draw [very thick, black, step=4cm,xshift=0cm, yshift=0cm] (0,0) grid +(8,4);
\draw[->,blue,thick,xshift=0.25cm,yshift=-0.25cm, rounded corners]
(0.0,4.0) -- (3.5,4.0) --
(0.0,3.5) -- (3.5,3.5) --
(0.0,3.0) -- (3.5,3.0) --
(0.0,2.5) -- (3.5,2.5) --
(0.0,2.0) -- (3.5,2.0) --
(0.0,1.5) -- (3.5,1.5) --
(0.0,1.0) -- (3.5,1.0) --
(0.0,0.5) -- (3.5,0.5) --
(4.0,4.0) -- (7.5,4.0) --
(4.0,3.5) -- (7.5,3.5) --
(4.0,3.0) -- (7.5,3.0) --
(4.0,2.5) -- (7.5,2.5) --
(4.0,2.0) -- (7.5,2.0) --
(4.0,1.5) -- (7.5,1.5) --
(4.0,1.0) -- (7.5,1.0) --
(4.0,0.5) -- (7.5,0.5);
\end{tikzpicture}
\label{fig:bufferindexing:a}
}
\quad
\subfigure[Pixel-based pattern]
{
\begin{tikzpicture}[scale=0.6,every node/.style={scale=0.75}]
\foreach \i in {0,...,15} {
\draw [very thin,black!40,text=black] (\i/2,0) -- (\i/2,4) node [above] at (\i/2+0.25,4) {$\i$};}
\foreach \i in {0,...,7} {
\draw [very thin,black!40,text=black] (0,\i/2) -- (8,\i/2) node [left] at (0,4-\i/2-0.25) {$\i$};}
\draw [very thick, black, step=4cm,xshift=0cm, yshift=0cm] (0,0) grid +(8,4);
\draw[->,blue,thick,xshift=0.25cm,yshift=-0.25cm, rounded corners]
(0.0,4.0) -- (7.5,4.0) --
(0.0,3.5) -- (7.5,3.5) --
(0.0,3.0) -- (7.5,3.0) --
(0.0,2.5) -- (7.5,2.5) --
(0.0,2.0) -- (7.5,2.0) --
(0.0,1.5) -- (7.5,1.5) --
(0.0,1.0) -- (7.5,1.0) --
(0.0,0.5) -- (7.5,0.5);
\end{tikzpicture}
\label{fig:bufferindexing:b}
}
\caption{Buffer layouts of an 16x8 image \protect\subref{fig:bufferindexing:a} before color conversion and \protect\subref{fig:bufferindexing:b} after color conversion.
The blue line indicates the access pattern.
\label{fig:bufferindexing}}
\end{center}
\end{figure}
The final output of the image is represented in interleaved RGB color space.
Each R, G and B value is represented as an unsigned character. Similar to
IDCT, vectorization in groups of four elements can be applied. In NVIDIA's
device compute capability 2.x or higher, global memory write instructions
support 1, 2, 4, 8 or 16 bytes~\cite{OpenCLProgrammingGuide4.2}. However, a
pixel consists of three bytes. Therefore, a work-item should perform
color conversion on a multiple of four pixels. An eight-pixel row has 24
elements. We group for pixels to six vectors of four elements as shown in
Figure~\ref{fig:vectorizingRGB}. The number of transfers is thereby
reduced by a factor of four.
\begin{figure}[ht]
\begin{center}
\begin{tikzpicture}[scale=0.7,every node/.style={scale=0.8}]
\foreach \j [count=\m from 0] in {0,0.5,...,3.5}
{
\draw [very thin,black!20,text=black,fill=red, opacity=0.1,text opacity=1]
(\j*3,2) rectangle (\j*3+0.5,2.5)
node [above] at (\j*3+0.25,2) {R};
\draw [very thin,black!20,text=black,fill=green, opacity=0.1,text opacity=1]
(\j*3+0.5,2) rectangle (\j*3+1,2.5)
node [above] at (\j*3+0.75,2) {G};
\draw [very thin,black!20,text=black,fill=blue, opacity=0.1,text opacity=1]
(\j*3+1,2) rectangle (\j*3+1.5,2.5)
node [above] at (\j*3+1.25,2) {B};
\draw node [above] at (\j*3+0.75,2.5) {pixel~\m};
}
\draw [densely dotted, black, step=1.5cm,yshift=2cm] (0,0) grid +(12,0.5);
\draw [densely dotted, black] (12,2.5) -- (0,2.5);
\foreach \j in {0,0.5,...,3.5}
{
\draw [very thin,black!20,text=black,fill=red, opacity=0.1,text opacity=1]
(\j*3,0) rectangle (\j*3+0.5,0.5)
node [above] at (\j*3+0.25,0) {R};
\draw [very thin,black!20,text=black,fill=green, opacity=0.1,text opacity=1]
(\j*3+0.5,0) rectangle (\j*3+1,0.5)
node [above] at (\j*3+0.75,0) {G};
\draw [very thin,black!20,text=black,fill=blue, opacity=0.1,text opacity=1]
(\j*3+1,0) rectangle (\j*3+1.5,0.5)
node [above] at (\j*3+1.25,0) {B};
}
\foreach \i in {0,...,5}
{
\draw node [above] at (\i*2+1,0.5) {vector~\i};
}
\draw [very thick, black, step=2cm] (0,0) grid +(12,0.5);
\draw [very thick, black] (0,0.5) -- (12,0.5);
\coordinate (a) at (6,1.75);
\coordinate (b) at (6,1);
\draw[->, >=latex, black, line width=3pt] (a) to node{} (b) node [right] at (6.1,1.5) {\textbf{vectorize}} ;
\end{tikzpicture}
\caption{Vectorization of interleaving RGB performed by one work-item in order to reduce global memory writes.
\label{fig:vectorizingRGB}}
\end{center}
\end{figure}
\vskip+5mm
\subsection{GPU Kernel Merging}
\label{KernelMerging}
Previously stored data in local memory is no longer accessible on the next
kernel invocation. Intermediate results must be stored back to global
memory at the end of each kernel invocation, which generates
unnecessary memory traffic.
Because the computation of color conversion has no data
dependency among pixels, it can be merged with the preceding kernel to reduce
global memory accesses.
An image with 4:4:4 subsampling does not require upsampling. The color
conversion kernel is merged with the IDCT kernel. Color conversion requires
information from all color spaces. Therefore, the IDCT kernel repeats the
computation three times for the three color spaces. At the end of IDCT, a work-item
holds Y, Cb and Cr rows in its registers. The work-item immediately performs
color conversion on the row without additional communication with
other work-items. Although a single work-item now performs three times more
IDCT computations, the storing of intermediate results in global memory
between the IDCT and color conversion kernel invocations are avoided.
With 4:2:2 subsampling, the color conversion and upsampling kernels are combined. We
use two OpenCL work-items to perform upsampling on a Cb and Cr row such that at
the end of upsampling, chrominance information of one row is stored in the
registers of each work-item. Only a row of Y space of the corresponding pixels
is loaded from global memory before starting color conversion. Our work-group
in the merged kernel, consisting of 128 work-items, processes two groups of
four blocks. Sixteen work-items are allocated per block, and 64 work-items
compute upsampling on the same index of different eight-pixel row segments to
avoid branch-divergence. At the end of upsampling, this work-group produces
sixteen image blocks.
We considered merging IDCT, upsampling and color conversion into one kernel.
Nevertheless, combining all kernels is not favourable because the number
of available registers constrains
the number of active work-groups per multiprocessor.
\subsection{Pipelined Kernel Execution}
\label{subsec:pipelinedGPU}
We observed that Huffman decoding consumed around half of the overall execution
time with the SIMD-version of libjpeg-turbo. Huffman decoding is sequential
and thus performed exclusively on the CPU. Subsequent decoding
steps, i.e., IDCT, upsampling and color conversions are highly
data-parallel and thus allocated to the GPU. In the following,
we refer to those steps as the {\em parallel part\/} of JPEG decoding.
In the execution model explained so
far, GPU computations are delayed until decoded entropy data becomes available,
as shown in Figure~\ref{fig:timelineGPU:a}. Because the GPU is un-utilized
during Huffman decoding, potential speedup is lost. Using the fact that
entropy data is decoded in order, GPU computations can start after sufficient
image rows have been decoded. Hence, Huffman decoding and GPU kernel execution
can be executed in a pipelined fashion, where the first pipeline stage, i.e.,
Huffman decoding, is executed on the CPU, and the second pipeline
stage, i.e., the parallel part, is executed on the GPU.
Figure~\ref{fig:timelineGPU:b} shows
the timeline of our pipelined execution mode.
An image is sliced horizontally into several chunks. As soon as the first
chunk is entropy decoded, the CPU transfers the data to the GPU. All OpenCL
commands are executed asynchronously. Hence, the CPU can resume Huffman
decoding immediately for the second chunk. The execution time of the GPU
kernel may not match the Huffman decoding time, because Huffman decoding varies
greatly with the image details contained in a chunk. If the GPU is faster than
the CPU, GPU computations are hidden behind Huffman-decoding on the CPU. The
overall execution time then consists of the Huffman decoding time of the entire
image plus the kernel invocation for the last portion of the image.
The most efficient chunk size is determined through static profiling on large
images. Chunk sizes are varied from the full height down to an eight pixel
stripe. The decoding speed tends to be faster as the number of chunks increases.
However, as chunks become too small, GPU utilization becomes low.
The best sizes from each image are selected. The final
partition size is chosen as the largest size on the best list to prevent from
choosing a size that is too small wrt.~GPU utilization.
\begin{figure}[htb]
\begin{center}
\subfigure[Normal GPU execution]{
\begin{tikzpicture}[
every node/.style={font=\normalsize,
minimum height=0.30cm,minimum width=0.15cm},
box/.style={minimum height=0.30cm,minimum width=1.25cm},
timelabel/.style={font=\normalsize,midway, above, sloped},
scale=0.75,every node/.style={scale=0.75},
]
\node [matrix, ampersand replacement=\&, very thin,column sep=0.15cm,row sep=0.30cm] (matrix) at (0,0) {
\node(0,0) (duration) {}; \& \node(0,0) (CPU) {}; \& \& \& \& \& \&
\node(0,0) (GPU) {}; \& \node(0,0) (GPU time) {}; \\
\node(0,0) (CPU time 0) {}; \& \node(0,0) (CPU 0) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 0) [box] {}; \& \node(0,0) (GPU time 0) {}; \\
\node(0,0) (CPU time 1) {}; \& \node(0,0) (CPU 1) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 1) [box] {}; \& \node(0,0) (GPU time 1) {}; \\
\node(0,0) (CPU time 2) {}; \& \node(0,0) (CPU 2) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 2) [box] {}; \& \node(0,0) (GPU time 2) {}; \\
\node(0,0) (CPU time 3) {}; \& \node(0,0) (CPU 3) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 3) [box] {}; \& \node(0,0) (GPU time 3) {}; \\
\node(0,0) (CPU time 4) {}; \& \node(0,0) (CPU 4) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 4) [box] {}; \& \node(0,0) (GPU time 4) {}; \\
\node(0,0) (CPU time 5) {}; \& \node(0,0) (CPU 5) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 5) [box] {}; \& \node(0,0) (GPU time 5) {}; \\
\node(0,0) (CPU time 6) {}; \& \node(0,0) (CPU 6) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 6) [box] {}; \& \node(0,0) (GPU time 6) {}; \\
\node(0,0) (CPU time 7) {}; \& \node(0,0) (CPU 7) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 7) [box] {}; \& \node(0,0) (GPU time 7) {}; \\
\node(0,0) (CPU time 8) {}; \& \node(0,0) (CPU 8) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 8) [box] {}; \& \node(0,0) (GPU time 8) {}; \\
\node(0,0) (CPU time 9) {}; \& \node(0,0) (CPU 9) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 9) [box] {}; \& \node(0,0) (GPU time 9) {}; \\
\node(0,0) (CPU time 10) {}; \& \node(0,0) (CPU 10) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 10) [box] {}; \& \node(0,0) (GPU time 10) {}; \\
\node(0,0) (CPU time 11) {}; \& \node(0,0) (CPU 11) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 11) [box] {}; \& \node(0,0) (GPU time 11) {}; \\
};
\fill
(CPU) node[draw,fill=white] (t1) {CPU}
(GPU) node[draw,fill=white] (t2) {GPU};
\draw [dashed] (t1.south) -- (CPU 11.south);
\draw [dashed] (t2.south) -- (GPU 11.south);
\filldraw[fill=blue!30]
(CPU 0.north west) rectangle (CPU 5.south east) node (huffman) [midway,align=center] {Huffman};
\filldraw[fill=blue!20]
(CPU 5.south west) rectangle (CPU 6.south east) node (dispatch) [midway,align=center] {Dispatch};
\filldraw[fill=green!20]
(GPU 5.south west) rectangle (GPU 11.south east) node (opencl) [midway,align=center] {GPU};
\draw [-latex] (dispatch.east) -- (dispatch.east -| GPU 4.west) node [timelabel] (TextNode) {Write};
\end{tikzpicture}
\label{fig:timelineGPU:a}}
\quad
\subfigure[Pipeline execution]{
\begin{tikzpicture}[
every node/.style={font=\normalsize,
minimum height=0.30cm,minimum width=0.15cm},
box/.style={minimum height=0.30cm,minimum width=1.6cm},
timelabel/.style={font=\normalsize,midway, above, sloped},
scale=0.75,every node/.style={scale=0.75},
]
\node [matrix, ampersand replacement=\&, very thin,column sep=0.15cm,row sep=0.30cm] (matrix) at (0,0) {
\node(0,0) (duration) {}; \& \node(0,0) (CPU) {}; \& \& \& \& \& \& \node(0,0) (GPU) {}; \& \node(0,0) (GPU time) {}; \\
\node(0,0) (CPU time 0) {}; \& \node(0,0) (CPU 0) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 0) [box] {}; \& \node(0,0) (GPU time 0) {}; \\
\node(0,0) (CPU time 1) {}; \& \node(0,0) (CPU 1) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 1) [box] {}; \& \node(0,0) (GPU time 1) {}; \\
\node(0,0) (CPU time 2) {}; \& \node(0,0) (CPU 2) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 2) [box] {}; \& \node(0,0) (GPU time 2) {}; \\
\node(0,0) (CPU time 3) {}; \& \node(0,0) (CPU 3) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 3) [box] {}; \& \node(0,0) (GPU time 3) {}; \\
\node(0,0) (CPU time 4) {}; \& \node(0,0) (CPU 4) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 4) [box] {}; \& \node(0,0) (GPU time 4) {}; \\
\node(0,0) (CPU time 5) {}; \& \node(0,0) (CPU 5) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 5) [box] {}; \& \node(0,0) (GPU time 5) {}; \\
\node(0,0) (CPU time 6) {}; \& \node(0,0) (CPU 6) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 6) [box] {}; \& \node(0,0) (GPU time 6) {}; \\
\node(0,0) (CPU time 7) {}; \& \node(0,0) (CPU 7) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 7) [box] {}; \& \node(0,0) (GPU time 7) {}; \\
\node(0,0) (CPU time 8) {}; \& \node(0,0) (CPU 8) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 8) [box] {}; \& \node(0,0) (GPU time 8) {}; \\
\node(0,0) (CPU time 9) {}; \& \node(0,0) (CPU 9) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 9) [box] {}; \& \node(0,0) (GPU time 9) {}; \\
\node(0,0) (CPU time 10) {}; \& \node(0,0) (CPU 10) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 10) [box] {}; \& \node(0,0) (GPU time 10) {}; \\
\node(0,0) (CPU time 11) {}; \& \node(0,0) (CPU 11) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 11) [box] {}; \& \node(0,0) (GPU time 11) {}; \\
};
\fill
(CPU) node[draw,fill=white] (t1) {CPU}
(GPU) node[draw,fill=white] (t2) {GPU};
\draw [dashed] (t1.south) -- (CPU 11.south);
\draw [dashed] (t2.south) -- (GPU 11.south);
\filldraw[fill=blue!30]
(CPU 0.north west) rectangle (CPU 2.north east)
node (huffman) [midway,align=center] {Huffman 1\\(to GPU)};
\filldraw[fill=blue!30]
(CPU 2.south west) rectangle (CPU 3.south east)
node (huffman) [midway,align=center] {Huffman 2\\(to GPU)};
\filldraw[fill=blue!30]
(CPU 4.north west) rectangle (CPU 6.south east)
node (huffman) [midway,align=center] {Huffman 3\\(to GPU)};
\filldraw[fill=blue!20]
(CPU 2.north west) rectangle (CPU 2.south east)
node (dispatch0) [text width=1.4cm, midway,align=center] {Dispatch};
\filldraw[fill=blue!20]
(CPU 3.south west) rectangle (CPU 4.north east)
node (dispatch1) [text width=1.4cm, midway,align=center] {Dispatch};
\filldraw[fill=blue!20]
(CPU 6.south west) rectangle (CPU 7.north east)
node (dispatch2) [text width=1.4cm, midway,align=center] {Dispatch};
\filldraw[fill=green!20]
(GPU 2.north west) rectangle (GPU 4.north east)
node (opencl0) [midway,align=center] {GPU\\ (Huffman 1)};
\filldraw[fill=green!20]
(GPU 4.north west) rectangle (GPU 6.north east)
node (opencl1) [midway,align=center] {GPU\\ (Huffman 2)};
\filldraw[fill=green!20]
(GPU 6.south west) rectangle (GPU 8.south east)
node (opencl2) [midway,align=center] {GPU\\ (Huffman 3)};
\draw [-latex] (dispatch0.east) -- (dispatch0.east -| GPU 2.west) node [timelabel] (TextNode 0) {Write};
\draw [-latex] (dispatch1.east) -- (GPU 4.west) node [timelabel] (TextNode 1) {Write};
\draw [-latex] (dispatch2.east) -- (dispatch2.east -| GPU 8.west) node [timelabel] (TextNode 2) {Write};
\end{tikzpicture}
\label{fig:timelineGPU:b}}
\caption{JPEG decompression timelines: \protect\subref{fig:timelineGPU:a}
GPU execution of the parallel part after Huffman decoding and
\protect\subref{fig:timelineGPU:b} pipelined execution of Huffman decoding and
GPU computations. The CPU reads back the results at the end of each kernel
invocation. The read arrows have been omitted for clarity.
\label{fig:timelineGPU}}
\end{center}
\end{figure}
\section{Heterogeneous JPEG Decoding}
\label{sec:heterogjpeg}
Low-end GPUs may be incapable of out-performing high-end CPUs. For such
CPU-GPU combinations, distributing the entire workload between GPU and CPU is
required. We propose a performance model and partitioning scheme that
dynamically balances the workload on a CPU-GPU system. We model execution time
based on an off-line profiling step. This profiling is required only once for
a given CPU-GPU combination. For profiling, we execute an instrumented version
of the JPEG decoder to determine the execution times of each decoding step for
a training set of images. Multivariate polynomial regression analysis is
applied to derive closed forms that characterize the performance of a given
CPU-GPU combination. We identified image entropy and the image dimensions as
the sole parameters for our performance model. At run-time, the closed forms
are evaluated for a given image to estimate execution times and load-balance
the decoding workload between the CPU and the GPU.
\subsection{Performance Model}
\label{subsec:perfModel}
Our training set consists of twelve images from an online image
benchmark~\cite{RawzorImageBenchmark} and seven self-taken images. Polynomial
regression poorly estimates performance for images with the dimensions outside
of the training set range. Thus, the training-set baseline images are cropped
to create combinations of width and height up to
25~megapixels. The total number of images in the training set
is 4449.
We categorize JPEG decoding stages into two phases: a sequential phase (Huffman
decoding) and a parallel phase (dequantization, IDCT, upsampling and color
conversion). The sequential phase is executed exclusively on CPU while the
parallel phase can be executed on either the CPU or the GPU. Execution times
are collected for four decoding modes: sequential, SIMD, GPU and
pipelined GPU. Execution time is measured using CPU timestamp counter
registers and the OpenCL event profiler.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{kernel}
\caption{Execution time of SIMD and GPU of the parallel phase on GTX 560 scales linearly as image size increased.
The other tested platforms showed a similar trend.
\label{fig:keneltime}}
\end{center}
\end{figure}
Figure~\ref{fig:keneltime} indicates that the parallel phase scales linearly
with respect to image size. Thus, we perform polynomial fits on the parallel
phase, $\ensuremath{P_{\mathit{CPU}}}$ and $\ensuremath{P_{\mathit{GPU}}}$, as a function of image width and height.
Huffman decoding does not show a linear relationship with image dimension.
We have observed that it varies on the complexity of the chrominance and
luminance of an image, which reflects on entropy size.
Figure~\ref{fig:huffmanExecution} suggests a linear relationship between
Huffman decoding time per pixel and entropy density. Because the encoded bitstream
occupies the largest portion of a JPEG file, the density of entropy coded data
can be approximated from image file size and image dimensions as
\begin{equation}
d = \frac{\mathit{ImageFileSize}}{w*h},
\end{equation}
where $w$ is the image width, $h$ is the image height and $d$ is the image's
entropy density per pixel. We model the Huffman decoding rate,
$T_{\mathit{HuffmanPerPixel}}$, using polynomial regression as a function of
entropy density. The Huffman decoding time of the entire image, $\ensuremath{T_{\mathit{Huff}}}$, is
approximated as follows.
\begin{equation}
\label{equation:huffman}
\ensuremath{T_{\mathit{Huff}}}(w,h,d) = T_{\mathit{HuffmanPerPixel}}(d) * w * h
\end{equation}
This equation assumes that entropy data is evenly distributed across an image, which
we found to be a workable approximation.
The variables to our performance model are image width, height and entropy data
size. We model each phase using polynomial regression up to a degree of seven.
The best fit model is selected by comparing Akaike information
criteria~\cite{Akaike19813}. Modelling with higher degrees is computationally
possible. However, we have observed that higher degrees do not imply a more
precise model, and performance may suffer from the higher prediction time
required to evaluate polynomials of higher degrees.
Evaluating polynomials of high degrees at run-time showed a noticeable negative
impact on the performance of the JPEG decoder. We rearranged all polynomials
in Horner form~\cite{horner1819} to reduce the number of multiplications
required for polynomial evaluations. With this optimization the prediction
overhead became negligible compared to the overall execution time for decoding.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{huffman}
\end{center}
\caption{Huffman decoding rate on GTX 560 with respect to the density of entropy in bytes per pixel along with best-fit lines.}
\label{fig:huffmanExecution}
\end{figure}
The overall execution time on a CPU can be expressed as a summation of the sequential phase and the parallel phase:
\begin{equation}
\ensuremath{T_{\mathit{total}}} = \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{CPU}}}(w,h),
\end{equation}
When we profile execution times on the GPU, OpenCL work-group sizes are
alternated from 4 MCUs to 32 MCUs to find the best work-group size for a
specific platform. Similar to the CPU model, the total execution time for the
GPU mode is expressed a summation of Huffman decoding time and GPU execution
time.
\begin{equation}
\ensuremath{T_{\mathit{total}}} = \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{GPU}}}(w,h)
\end{equation}
Data transfers between CPU and GPU device generate significant
overhead~\cite{GPUtransfer2011}. The time collected for the GPU includes
data transfer overhead and
the GPU kernel computation,
\begin{equation}
\label{gpubeakdown}
\ensuremath{P_{\mathit{GPU}}}(w,h) = O_{w}(w,h) + T_{kernel}(w,h) + O_{r}(w,h),
\end{equation}
where $O_{w}$ and $O_{r}$ are data transfer costs from the CPU to the
GPU and vice versa.
The input and output buffers are pinned, to achieve faster transfers~\cite{OpenCLProgrammingGuide4.2}.
\begin{figure*}[htb]
\begin{center}
\subfigure[SPS]{
\begin{tikzpicture}[
every node/.style={font=\normalsize,
minimum height=0.30cm,minimum width=0.3cm},
box/.style={minimum height=0.30cm,minimum width=2.1cm},
timelabel/.style={font=\normalsize,midway, above, sloped},
scale=0.75,every node/.style={scale=0.72},
]
\node [matrix, ampersand replacement=\&, very thin,column sep=0.17cm,row sep=0.30cm] (matrix) at (0,0) {
\node(0,0) (duration) {}; \& \node(0,0) (CPU) {}; \& \& \& \& \& \&
\node(0,0) (GPU) {}; \& \node(0,0) (GPU time) {}; \\
\node(0,0) (CPU time 0) {}; \& \node(0,0) (CPU 0) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 0) [box] {}; \& \node(0,0) (GPU time 0) {}; \\
\node(0,0) (CPU time 1) {}; \& \node(0,0) (CPU 1) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 1) [box] {}; \& \node(0,0) (GPU time 1) {}; \\
\node(0,0) (CPU time 2) {}; \& \node(0,0) (CPU 2) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 2) [box] {}; \& \node(0,0) (GPU time 2) {}; \\
\node(0,0) (CPU time 3) {}; \& \node(0,0) (CPU 3) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 3) [box] {}; \& \node(0,0) (GPU time 3) {}; \\
\node(0,0) (CPU time 4) {}; \& \node(0,0) (CPU 4) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 4) [box] {}; \& \node(0,0) (GPU time 4) {}; \\
\node(0,0) (CPU time 5) {}; \& \node(0,0) (CPU 5) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 5) [box] {}; \& \node(0,0) (GPU time 5) {}; \\
\node(0,0) (CPU time 6) {}; \& \node(0,0) (CPU 6) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 6) [box] {}; \& \node(0,0) (GPU time 6) {}; \\
\node(0,0) (CPU time 7) {}; \& \node(0,0) (CPU 7) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 7) [box] {}; \& \node(0,0) (GPU time 7) {}; \\
\node(0,0) (CPU time 8) {}; \& \node(0,0) (CPU 8) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 8) [box] {}; \& \node(0,0) (GPU time 8) {}; \\
\node(0,0) (CPU time 9) {}; \& \node(0,0) (CPU 9) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 9) [box] {}; \& \node(0,0) (GPU time 9) {}; \\
\node(0,0) (CPU time 10) {}; \& \node(0,0) (CPU 10) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 10) [box] {}; \& \node(0,0) (GPU time 10) {}; \\
\node(0,0) (CPU time 11) {}; \& \node(0,0) (CPU 11) [box] {}; \& \& \& \& \& \&
\node(0,0) (GPU 11) [box] {}; \& \node(0,0) (GPU time 11) {}; \\
};
\fill
(CPU) node[draw,fill=white] (t1) {CPU}
(GPU) node[draw,fill=white] (t2) {GPU};
\draw [dashed] (t1.south) -- (CPU 11.south);
\draw [dashed] (t2.south) -- (GPU 11.south);
\filldraw[fill=blue!30]
(CPU 0.north west) rectangle (CPU 5.south east) node (huffman) [midway,align=center] {Huffman};
\filldraw[fill=blue!20]
(CPU 5.south west) rectangle (CPU 6.south east) node (dispatch) [text width=2.0cm, midway,align=center] {Dispatch};
\filldraw[fill=green!20]
(GPU 5.south west) rectangle (GPU 10.north east) node (opencl) [midway,align=center] {GPU};
\filldraw[fill=red!10]
(CPU 6.south west) rectangle (CPU 10.north east) node (simd)
[midway,align=center] {SIMD};
\draw[<->,blue] (CPU time 5.south) -- (CPU time 0.north)
node [timelabel] (TextNode 0) {$\ensuremath{T_{\mathit{Huff}}}$};
\draw[<->,blue] (CPU time 6.south) -- (CPU time 5.south)
node [timelabel] (TextNode 1) {$\ensuremath{T_{\mathit{disp}}}$};
\draw[<->,blue] (GPU time 10.north) -- (GPU time 5.south)
node [timelabel,below] (TextNode 2) {$\ensuremath{P_{\mathit{GPU}}}$};
\draw[<->,blue] (CPU time 10.north) -- (CPU time 6.south)
node [timelabel] (TextNode 1) {$\ensuremath{P_{\mathit{CPU}}}$};
\draw [-latex] (dispatch.east) -- (dispatch.east -| GPU 4.west) node [timelabel] (TextNode) {Write};
\end{tikzpicture}
\label{fig:timelineHeterog:a}
}
\qquad\quad
\subfigure[Overlapped]{
\begin{tikzpicture}[
every node/.style={font=\normalsize,
minimum height=0.30cm,minimum width=0.3cm},
box/.style={minimum height=0.30cm,minimum width=2.1cm},
timelabel/.style={font=\normalsize,midway, above, sloped},
scale=0.75,every node/.style={scale=0.72},
]
\node [matrix, ampersand replacement=\&, very thin,column sep=0.17cm,row sep=0.30cm] (matrix) at (0,0) {
\node(0,0) (CPU) {}; \& \& \& \& \& \&
\node(0,0) (GPU) {}; \& \node(0,0) (GPU time) {}; \\
\node(0,0) (CPU 0) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 0) [box] {};\\
\node(0,0) (CPU 1) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 1) [box] {};\\
\node(0,0) (CPU 2) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 2) [box] {};\\
\node(0,0) (CPU 3) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 3) [box] {};\\
\node(0,0) (CPU 4) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 4) [box] {};\\
\node(0,0) (CPU 5) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 5) [box] {};\\
\node(0,0) (CPU 6) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 6) [box] {};\\
\node(0,0) (CPU 7) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 7) [box] {};\\
\node(0,0) (CPU 8) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 8) [box] {};\\
\node(0,0) (CPU 9) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 9) [box] {};\\
\node(0,0) (CPU 10) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 10) [box] {};\\
\node(0,0) (CPU 11) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 11) [box] {};\\
};
\fill
(CPU) node[draw,fill=white] (t1) {CPU}
(GPU) node[draw,fill=white] (t2) {GPU};
\draw [dashed] (t1.south) -- (CPU 11.south);
\draw [dashed] (t2.south) -- (GPU 11.south);
\filldraw[fill=blue!30]
(CPU 0.north west) rectangle (CPU 4.north east) node (huffman) [text width=2.0cm, midway,align=center] {Huffman 1\\ (to GPU)};
\filldraw[fill=blue!30]
(CPU 5.north west) rectangle (CPU 6.south east) node (huffman) [text width=2.0cm, midway,align=center] {Huffman 2\\ (for CPU)};
\filldraw[fill=blue!20]
(CPU 4.north west) rectangle (CPU 5.north east) node (dispatch) [text width=2.0cm, midway,align=center] {Dispatch};
\filldraw[fill=green!20]
(GPU 4.north west) rectangle (GPU 8.south east) node (opencl) [text width=2.0cm, midway,align=center] {GPU};
\filldraw[fill=red!10]
(CPU 6.south west) rectangle (CPU 8.south east) node (simd)
[midway,align=center] {SIMD};
\draw [-latex] (dispatch.east) -- (dispatch.east -| GPU 4.west) node [timelabel] (TextNode) {Write};
\end{tikzpicture}
\label{fig:timelineHeterog:b}}
\qquad\quad
\subfigure[PPS]{
\begin{tikzpicture}[
every node/.style={font=\normalsize,
minimum height=0.30cm,minimum width=0.3cm},
box/.style={minimum height=0.30cm,minimum width=2.1cm},
timelabel/.style={font=\normalsize,midway, above, sloped},
scale=0.75,every node/.style={scale=0.72},
]
\node [matrix,ampersand replacement=\&, very thin,column sep=0.17cm,row sep=0.30cm] (matrix) at (0,0) {
\node(0,0) (CPU) {}; \& \& \& \& \& \& \node(0,0) (GPU) {}; \&\\
\node(0,0) (CPU 0) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 0) [box] {};\\
\node(0,0) (CPU 1) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 1) [box] {};\\
\node(0,0) (CPU 2) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 2) [box] {};\\
\node(0,0) (CPU 3) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 3) [box] {};\\
\node(0,0) (CPU 4) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 4) [box] {};\\
\node(0,0) (CPU 5) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 5) [box] {};\\
\node(0,0) (CPU 6) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 6) [box] {};\\
\node(0,0) (CPU 7) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 7) [box] {};\\
\node(0,0) (CPU 8) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 8) [box] {};\\
\node(0,0) (CPU 9) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 9) [box] {};\\
\node(0,0) (CPU 10) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 10) [box] {};\\
\node(0,0) (CPU 11) [box] {}; \& \& \& \& \& \& \node(0,0) (GPU 11) [box] {};\\
};
\fill
(CPU) node[draw,fill=white] (t1) {CPU}
(GPU) node[draw,fill=white] (t2) {GPU};
\draw [dashed] (t1.south) -- (CPU 11.south);
\draw [dashed] (t2.south) -- (GPU 11.south);
\filldraw[fill=blue!30]
(CPU 0.north west) rectangle (CPU 2.north east)
node (huffman) [midway,align=center] {Huffman 1\\ (to GPU)};
\filldraw[fill=blue!30]
(CPU 2.south west) rectangle (CPU 3.south east)
node (huffman) [midway,align=center] {Huffman 2\\ (to GPU)};
\filldraw[fill=blue!30]
(CPU 4.north west) rectangle (CPU 6.north east)
node (huffman) [midway,align=center] {Huffman 3\\ (to GPU)};
\filldraw[fill=blue!30]
(CPU 6.south west) rectangle (CPU 7.north east)
node (huffman) [midway,align=center] {Huffman (CPU)};
\filldraw[fill=blue!20]
(CPU 2.north west) rectangle (CPU 2.south east)
node (dispatch0) [text width=2.0cm,midway,align=center] {Dispatch};
\filldraw[fill=blue!20]
(CPU 3.south west) rectangle (CPU 4.north east)
node (dispatch1) [text width=2.0cm,midway,align=center] {Dispatch};
\filldraw[fill=blue!20]
(CPU 6.north west) rectangle (CPU 6.south east)
node (dispatch2) [text width=2.0cm,midway,align=center] {Dispatch};
\filldraw[fill=green!20]
(GPU 2.north west) rectangle (GPU 4.north east)
node (opencl0) [midway,align=center] {GPU\\ (Huffman 1)};
\filldraw[fill=green!20]
(GPU 4.north west) rectangle (GPU 6.north east)
node (opencl1) [midway,align=center] {GPU\\ (Huffman 2)};
\filldraw[fill=green!20]
(GPU 6.north west) rectangle (GPU 7.south east)
node (opencl2) [midway,align=center] {GPU\\ (Huffman 3)};
\filldraw[fill=red!10]
(CPU 7.north west) rectangle (CPU 7.south east) node (simd)
[midway,align=center] {SIMD};
\draw [-latex] (dispatch0.east) -- (dispatch0.east -| GPU 2.west) node [timelabel] (TextNode 0) {Write};
\draw [-latex] (dispatch1.east) -- (GPU 4.west) node [timelabel] (TextNode 1) {Write};
\draw [-latex] (dispatch2.east) -- (dispatch2.east -| GPU 6.west) node [timelabel] (TextNode 2) {Write};
\end{tikzpicture}
\label{fig:timelineHeterog:c}}
\end{center}
\vspace{-4mm}
\caption{Heterogeneous JPEG decoding timelines of three execution models:
\protect\subref{fig:timelineHeterog:a} SPS,
\protect\subref{fig:timelineHeterog:b} overlapped Huffman decoding and GPU
execution, and \protect\subref{fig:timelineHeterog:c} PPS.
Partitioning schemes are depicted proportionally; PPS achieves the highest
overlap of GPU kernel execution with the sequential Huffman decoding on the
CPU. Note that the last
GPU invocation is shorter because the work is shared with the CPU (SIMD).
\label{fig:timelineHeterog}}
\end{figure*}
\subsection{Partitioning Schemes}
For reasons introduced in Section~\ref{sec:intro}, Huffman decoding constitutes
the non-parallelizable part of JPEG decoding, which is thus entirely executed on
the CPU. Subsequent decoding steps, i.e., IDCT, upsampling and color conversion
constitute the parallelizable part for which we utilize both the CPU and the GPU.
For the parallelizable part, our partitioning scheme splits images horizontally such that the
initial $x$~rows of the image are assigned to the GPU, and the remaining
$h-x$~rows are assigned to the CPU. The value for variable~$x$ is chosen
such that the overall execution times for the CPU and GPU are equal, i.e., the load
is equally balanced. Variable~$x$ is rounded to the nearest value evenly divisible
by the number of rows in an MCU. This requirement is due to libjpeg-turbo's
convention to decode images in units of MCUs.
The input parameters to our partitioning schemes are the image dimensions and the
image entropy, approximated by bytes/pixel derived from the image data size and
the image dimensions.
\subsubsection{Simple Partitioning Scheme (SPS):}
The simplest approach is to parallelize the computations after Huffman
decoding. CPU and GPU perform the parallel phase concurrently.
Figure~\ref{fig:timelineHeterog:a} illustrates the SPS partitioning scheme.
The CPU first performs entropy decoding of the entire image, then partitions the resulting
image data in two parts. The first part is processed by the GPU and the second part by
the CPU. Data transfer commands between CPU and GPU and kernel launching
commands are asynchronous
calls. Hence, the CPU is allowed to continue execution after dispatching
commands to the GPU.
The overall execution time can be modelled as the maximum time of the two architectures.
\begin{equation}
\ensuremath{T_{\mathit{total}}} = max(\ensuremath{T_{\mathit{CPU}}},\ensuremath{T_{\mathit{GPU}}})
\end{equation}
With the SPS model, the CPU execution time, $\ensuremath{T_{\mathit{CPU}}}$, and the GPU execution time, $\ensuremath{T_{\mathit{GPU}}}$,
are expressed as
{\small
\begin{subequations}
\begin{align}
\ensuremath{T_{\mathit{CPU}}}(w,h) &= \ensuremath{T_{\mathit{Huff}}}(w,h,d) +,\ensuremath{T_{\mathit{disp}}}(w,h-x) + \ensuremath{P_{\mathit{CPU}}}(w,x), \text{ and}\\
\ensuremath{T_{\mathit{GPU}}}(w,h) &= \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{GPU}}}(w,h-x),
\end{align}
\end{subequations}
}
where $x$ is the number of rows assigned to the CPU, and $\ensuremath{T_{\mathit{disp}}}$ is the
amount
of time the CPU spends on the OpenCL kernel invocation. $\ensuremath{P_{\mathit{GPU}}}$ includes kernel
execution time and data transfer overhead. The workload is
considered well-balanced when the parallel parts on both architectures achieve
the same execution time.
{\small
\begin{equation}
f(x) = \ensuremath{T_{\mathit{disp}}}(w,h-x) + \ensuremath{P_{\mathit{CPU}}}(w,x) - \ensuremath{P_{\mathit{GPU}}}(w,h-x)\label{equation:simpleHeterog}
\end{equation}
}
The only unknown variable is the number of rows assigned to the CPU, i.e.,~$x$.
When $f(x)$ become zero, the execution time is balanced.
This problem is equivalent to the root solving problem.
At run-time, the root can be estimated using Newton's method,
\begin{equation}
x_{n+1} = x_n - \frac{f(x_n)}{f'(x_n)},
\end{equation}
where $x_n$ is the initial partitioning height to the CPU, $x_{n+1}$ is the new
height approximation, $f(x_n)$ is given as
Equation~\eqref{equation:simpleHeterog}, and $f'(x_n)$ is the first derivation.
Newton's method is performed recursively until no better partition can be
found. The GPU computes the parallel phase on the sub-image of size $w$ by
$h-x$, while the CPU computes the remaining $x$ image rows.
\subsubsection{Pipelined Partitioning Scheme (PPS):}
The GPU is underutilized during the Huffman decoding stage in SPS.
We have demonstrated in Section~\ref{subsec:pipelinedGPU} that sequential
Huffman decoding can be parallelized with GPU kernel execution. Entropy data
for the CPU can be decoded simultaneously with GPU computations as illustrated in
Figure~\ref{fig:timelineHeterog:b}. The concurrent execution happens after the
Huffman decoding part for the GPU. The total execution time on each architecture can be modelled as follows.
{\small
\begin{subequations}
\begin{align}
\ensuremath{T_{\mathit{CPU}}} &= \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{CPU}}}(w,x) + \ensuremath{T_{\mathit{disp}}}(w,h-x)\\
\ensuremath{T_{\mathit{GPU}}} &= \ensuremath{T_{\mathit{Huff}}}(w,h-x,d) + \ensuremath{P_{\mathit{GPU}}}(w,h-x)
\end{align}
\end{subequations}
}
This partitioning scheme balances GPU execution with the sum of OpenCL
dispatching time, entropy decoding time for the CPU part and the computation
time of the parallel part on the CPU.
{\small
\begin{equation}
\label{equation:overlap}
\begin{split}
f(x) = &\ensuremath{T_{\mathit{disp}}}(w,h-x) + \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{CPU}}}(w,x) \\
&- \ensuremath{P_{\mathit{GPU}}}(w,h-x)
\end{split}
\end{equation}
}
Similar to SPS, we use Newton's method to approximate variable~$x$ at run-time.
Further parallelism can be achieved by
pipelining the GPU kernel executions with this partitioning scheme.
Figure~\ref{fig:timelineHeterog:c} shows an
execution timeline of PPS. The parallel part for the first image chunk
(Huffman 1)
is started on the GPU immediately after it has been transferred from
the CPU.
{\small
\begin{subequations}
\begin{align}
\ensuremath{T_{\mathit{CPU}}} &= \ensuremath{T_{\mathit{Huff}}}(w,h,d) + \ensuremath{P_{\mathit{CPU}}}(w,x) + \ensuremath{T_{\mathit{disp}}}(w,h-x)\\
\ensuremath{T_{\mathit{GPU}}} &= \ensuremath{T_{\mathit{Huff}}}(w,c,d) + \ensuremath{P_{\mathit{GPU}}}(w,h-x)
\end{align}
\end{subequations}
}
In the above equation,
$c$ is the number of image rows per chunk decoded on the GPU.
The chunk size has been determined through profiling as explained in Section~\ref{subsec:pipelinedGPU}.
The number of rows per chunk can be calculated by chunk size divided by image width.
Therefore, the partitioning equation becomes
{\small
\begin{equation}
\begin{split}
f(x) = &\ensuremath{T_{\mathit{Huff}}}(w,h-c,d) + \ensuremath{P_{\mathit{CPU}}}(w,x) + \ensuremath{T_{\mathit{disp}}}(w,h-x)\\
&- \ensuremath{P_{\mathit{GPU}}}(w,h-x),
\end{split}
\end{equation}
}
where $h-c$ denotes the remaining rows in an image after the first chunk.
Our Huffman decoding time estimation model assumes a uniform entropy distribution
across an image. However, the density of entropy data is unlikely to be evenly
distributed in practice. Using the Huffman decoding time model to estimate the time
for a certain chunk is often imprecise because the average
density and the actual density of the chunk are mismatched. We compensate the
error by re-partitioning. Throughout the computation, we keep records of the
actual Huffman decoding times. Before entropy decoding of the last GPU chunk,
workload distribution is re-calculated.
At this point, one GPU chunk and the CPU partition remain unprocessed.
Thus, a modification of Equation~\eqref{equation:overlap} can be used,
{\small
\begin{equation}
\label{equation:re-calculation}
\begin{split}
f(x) = &\ensuremath{T_{\mathit{disp}}}(w,h'-x') + \ensuremath{T_{\mathit{Huff}}}(w,h',d') + \ensuremath{P_{\mathit{CPU}}}(w,x') \\
&- \ensuremath{P_{\mathit{GPU}}}(w,h-x) - P_{prevGPU},
\end{split}
\end{equation}
}
where $h'$ is the unprocessed height, $x'$ is the new height allocated to the GPU,
and $d'$ is the new density rate. The previous kernel execution may not
complete by the time of re-partitioning. $T_{prevGPU}$ is an estimated
remaining time from the previous GPU kernel invocation that potentially
influences the new partitioning scheme. The remaining height, $h'$, is known at
run-time. The only unknown variable to be solved at run-time, is $x'$.
We estimate the total Huffman decoding time using
Equation~\eqref{equation:huffman}, and the actual Huffman decoding time of
previous chunks are known at run-time. Figure~\ref{fig:huffmanExecution}
implies a linear proportional relationship between an image dimension and
Huffman decoding time. The new density is calculated using the ratio of the
remaining Huffman time and the image height.
\begin{equation}
d' = \frac{\mathit{HuffmanDecodingTimeRatio}}{\mathit{ImageHeightRatio}} * d,
\end{equation}
Above, $HuffmanDecodingTimeRatio$ is the ratio of the remaining decoding
time to the estimated total decoding time, and $ImageHeightRatio$ is the
ratio of unprocessed height to the total image height. When the ratio of the
remaining Huffman time is greater than the height ratio, the remaining part of
an image consists of more detail. It indicates that the entropy data rate
becomes denser and more workload should be allocated to the GPU. Otherwise, the
entropy data rate becomes less dense, and more workload should be allocated to
the CPU.
Even though Figure~\ref{fig:timelineHeterog:c} shows a small gain compared
to Figure~\ref{fig:timelineGPU:b}, this approach actually yields
large improvements on a hardware configuration where the CPU is more powerful
than the GPU.
\section{Experimental Results}
\label{sec:experiment}
We conducted an extensive experimental evaluation on six versions of the JPEG
decoder, namely the sequential version, SIMD, GPU, pipelined GPU, SPS and PPS,
on three representative platforms specified in Table~\ref{tab:spec}. All file
I/O instructions were disabled to minimize time variations that do not reflect
the actual performance of the algorithm. To demonstrate the effectiveness of
our implementation, we used two chroma subsamplings,
i.e.,
4:2:2 and 4:4:4. The other subsamplings are decoded in a similar manner as
4:2:2 images.
For the performance evaluation we used a new set of images that does not share
any images with the training set. Our image test-set consists of fourteen
images from {CorpusNielsFrohling} and three self-taken images. These images
are cropped to various sizes summing up to the total of 3597 images for each
subsampling.
\begin{table}[h!btp]
\centering
\small
\begin{tabular}{|l|c|c|c|}
\hline
\textbf{Machine name} & \textbf{GT 430} & \textbf{GTX 560} & \textbf{GTX 680}\\ \hline \hline
CPU model & Intel i7-2600k & Intel i7-2600k & Intel i7-3770k \\ \hline
CPU frequency & 3.4 GHz & 3.4 GHz & 3.5 GHz \\ \hline
No. of CPU cores & 4 & 4 & 4\\ \hline \hline
GPU model & \begin{tabular}[x]{@{}c@{}}NVIDIA\\GT 430\end{tabular}
& \begin{tabular}[x]{@{}c@{}}NVIDIA\\GTX 560Ti\end{tabular}
& \begin{tabular}[x]{@{}c@{}}NVIDIA\\GTX 680\end{tabular} \\ \hline
GPU core frequency & 700 MHz & 822MHz & 1006MHz \\ \hline
No. of GPU cores & 96 & 384 & 1536 \\ \hline
GPU memory size & 1024 MB & 1024 MB & 2048 MB \\ \hline
Compute Capability & 2.1 & 2.1 & 3.0 \\ \hline \hline
Ubuntu version & 11.04 & 12.04 & 11.04 \\ \hline
Linux Kernel & 2.6.38 & 3.5.0 & 2.6.38 \\ \hline
GCC version& 4.5.2 & 4.6.3 & 4.5.2 \\ \hline
\end{tabular}
\caption{Hardware Specifications.}
\label{tab:spec}
\end{table}
\subsection{OpenCL Kernel Execution}
Figure~\ref{fig:GPUBar} depicts a break-down of execution times
for the sequential CPU,
SIMD and GPU modes. The y-axis is normalized with respect to
SIMD execution times. Our GPU computations on all tested architectures surpass
the sequential CPU execution. For this specific image size, the GPU
computation was able to reduce 35.5\% and 40.8\% of the overall execution time
on GTX 560 and GTX 680 respectively. The kernel execution from IDCT to
color conversion was 10x faster than the SIMD execution on the GTX 560 and 13.7x
faster on the GTX 680. However, taking data transfer overhead into account, the
performance improvements were reduced to 2.6x and 4.3x.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{bar}
\caption{Decoding time normalized with respect to JPEG decompression in
SIMD mode. The decoded image's dimension is 2048x2048 with 4:2:2 subsampling.
Shown are the execution time break-downs of libjpeg-turbo's sequential JPEG decoder
on the CPU, SIMD
execution on a CPU with libjpeg-turbo, and our GPU execution.
\label{fig:GPUBar}}
\end{center}
\end{figure}
It follows from Fig.~\ref{fig:GPUBar}
that performance improvements are not guaranteed by migrating computations exclusively
to a GPU. GT~430, consisting of 96 cores, is the weakest GPU among the three
representative machines. The experimental result on the 2048x2048 image showed
a 23\% slow-down compared to SIMD execution on an Intel i7. The kernel
execution shows a 27\% slower data transfer between CPU and GPU.
\subsection{Heterogeneous JPEG Decoding Performance}
We evaluated our heterogeneous JPEG decompression models with respect to the
SIMD-version of libjpeg-turbo. Figure~\ref{fig:Speedup} shows the average speedups
with standard deviation bars as image size increased. Due to space
limitations, we only provide the results for 4:4:4 subsampling. A similar
trend was observed for 4:2:2 subsampling. Table~\ref{tab:speedup422} and
Table~\ref{tab:speedup444} summarize the performances of 4:2:2 and 4:4:4
subsampling respectively.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{speedup}
\caption{Average speedups over libjpeg-turbo's SIMD execution with respect
to image size in pixels on the three representative machines. The error bar
represents standard deviation. \label{fig:Speedup}}
\end{center}
\end{figure}
\begin{table}[h!tbp]
\centering
\small
\begin{tabular}{|l|c|c|c|}
\hline
\textbf{Mode} & \textbf{GT 430} & \textbf{GTX 560} & \textbf{GTX 680}\\ \hline
GPU & $0.72 \pm 5.35\%$ & $1.59 \pm 7.50\%$ & $1.94 \pm 12.94\%$ \\ \hline
Pipeline & $0.92 \pm 13.67\%$ & $2.19 \pm 20.21\%$ & $2.33 \pm 20.75\%$ \\ \hline
SPS & $1.31 \pm 9.54\%$ & $1.81 \pm 10.13\%$ & $2.04 \pm 15.15\%$ \\ \hline
PPS & $1.54 \pm 10.93\%$ & $2.34 \pm 15.19\%$ & $2.52 \pm 17.08\%$ \\ \hline
\end{tabular}
\caption{Average speedup and coefficient of variation over SIMD execution when decoding 4:2:2 subsampled images.}
\label{tab:speedup422}
\end{table}
\begin{table}[h!tbp]
\centering
\small
\begin{tabular}{|l|c|c|c|}
\hline
\textbf{Mode} & \textbf{GT 430} & \textbf{GTX 560} & \textbf{GTX 680}\\ \hline
GPU & $0.66 \pm 5.82\%$ & $1.49 \pm 5.87\%$ & $1.81 \pm 10.84\%$ \\ \hline
Pipeline & $0.83 \pm 13.48\%$ & $2.14 \pm 19.97\%$ & $2.26 \pm 19.48\%$ \\ \hline
SSP & $1.27 \pm 8.62\%$ & $1.76 \pm 8.12\%$ & $1.94 \pm 12.55\%$ \\ \hline
PPS & $1.50 \pm 10.46\%$ & $2.34 \pm 14.33\%$ & $2.45 \pm 15.02\%$ \\ \hline
\end{tabular}
\caption{Average speedup and coefficient of variation over SIMD execution when decoding 4:4:4 subsampled images.}
\label{tab:speedup444}
\end{table}
PPS achieves the highest performance on all machines. It attains average speedups
of 1.5x, 2.3x and 2.5x over SIMD mode and 3.1x, 4.8x and 5.2x over sequential
execution on GTX 430, GTX 560 and GTX 680 respectively. The highest-recorded
speedups were 4.2x faster than SIMD and 8.5x faster than sequential execution
on GTX 680. PPS does not show a signification improvement over pipelined GPU
execution on GTX 560 and GTX 680 because most GPU kernel executions were
sufficiently fast to hide within the Huffman decompression time. Therefore,
only a small amount of workload was allocated to the CPU, and a small improvement
was achieved.
On GT 430, the GPU mode and the pipelined GPU execution mode failed to surpass
SIMD. As a result, both of our partitioning schemes distributed the larger
partition to the CPU. Despite the slow GPU, the cooperative CPU-GPU
executions achieved speedups over libjpeg-turbo's SIMD mode.
The pipelined execution is always faster than a single large GPU kernel
invocation because entropy is decoded simultaneously with a GPU computation to
reduce the hardware idle time. When the decoded image has a size smaller than
the pre-determined chunk size, the image is executed as one GPU kernel invocation.
Therefore, no improvement is shown over the normal GPU mode.
It should be noted that the GTX 680 has larger coefficients of variation than the
other machines. This fluctuation reflected the contribution of Huffman
decoding time to the speedup calculation. An image with larger entropy data takes
longer time to decode. As a result, the overall speedup becomes smaller than an
image with sparser entropy. The faster GPU is more sensitive to the change.
Therefore, GTX 680 suffered the highest impact from a small change in Huffman
decoding time compared to the other tested machines.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{theoretical_speedup}
\caption{Speedup Comparison of PPS execution to the maximum achievable speedup on GTX 680.
The points represent the mean percent achieved along with standard deviation bars.
\label{fig:pcnt_achvd}}
\end{center}
\end{figure}
According to Amdahl's law, the maximum attainable speedup is restricted by
the sequential portion of the program. Equation~\eqref{equation:amdahl}
states the theoretical speedup assuming an infinite number of processors.
\begin{equation}
\label{equation:amdahl}
\mathit{Speedup} = \frac{\ensuremath{T_{\mathit{total}}}}{\ensuremath{T_{\mathit{total}}}*(1-P)}
\end{equation}
$P$ is the fraction of the parallelizable portion of the program, and $1-P$ is
the serial portion, which in this case, is entropy (Huffman) decoding.
Thus, the maximum achievable speedup over libjpeg-turbo's SIMD-version
can be written as
\begin{equation}
\mathit{Speedup} = \frac{\ensuremath{T_{\mathit{total}}}}{\ensuremath{T_{\mathit{Huff}}}},
\end{equation}
where $\ensuremath{T_{\mathit{total}}}$ is the decoding time of the SIMD-version.
We compared the speedup of our approach to the theoretically attainable
speedup in Figure~\ref{fig:pcnt_achvd}. PPS stabilizes at an average speedup
of 88\% and attains its peak at 95\% of the theoretically attainable speedup.
For small images, the speedup was slightly higher than half of the maximum
attainable speedup because these images were partitioned into few chunks for
pipeline execution.
Consequently, less work was executed in parallel with entropy decoding.
Increasing the number of chunks would result in a lack of data for GPU computation.
As image size increases, an image is split into more chunks, and thus, less work of the parallelizable phase is visible to the user.
\begin{figure}[htb]
\begin{center}
\includegraphics[clip=true]{prediction_error}
\caption{The average CPU and GPU execution time with standard deviation during parallel executions are balanced indicating balance workload between architectures.
\label{fig:Prediction}}
\end{center}
\end{figure}
Figure~\ref{fig:Prediction} shows execution times of the CPU and the GPU during
the parallel execution. In the simple heterogeneous execution, the entropy
decoding time was omitted from CPU time as it is sequentially executed on the GPU.
Similarly, the entropy decoding of the first image chunk of the pipelined
execution was omitted. GPU and CPU shared similar execution times indicating
well-balanced loads. The main contribution to the variation in the
pipelined heterogeneous mode is Huffman decoding time.
\section{Related Work}
\label{sec:related}
DCT and IDCT algorithms are computationally intensive, but exhibit a high potential for parallelism.
Various image processing algorithms, including DCT and IDCT, have been implemented for a GPU architecture by Yang et al.~\cite{yang2008parallel}.
The authors utilized CUDA and applied CUFFT~\cite{cufftGuideV5}, a CUDA fast Fourier transform (FFT) library, to perform DCT and IDCT.
However, extracting DCT from FFT introduces extra computational overhead.
The NVIDIA GPU Computing SDK provides DCT and IDCT sample code to demonstrate GPU programming.
The kernel's input and output data type are float.
In contrast to our implementation, the input buffer to our kernel is an array of short and the output buffer is an array of unsigned character.
These data types are vectorized to minimize global memory transfer overhead.
Moreover, we combine dequantization, IDCT and color conversion in a single kernel to reduce data communication.
A task-parallel implementation of JPEG decoding using libjpeg-turbo has been explored by Hong et al.~\cite{hong2012task}.
Fork/join parallelism is applied to decode an image on CPU cores simultaneously.
GPUJPEG~\cite{gpujpeg} is an open source JPEG image compression and decompression library for NVIDIA designed for real-time video.
Tasks for image decoding are divided between a CPU and GPUs where the CPU performs file I/O operations and Huffman decoding while the GPUs compute IDCT and color conversion.
The GPU kernel is, yet, optimized.
Although the computationally intensive tasks are parallelized on the GPUs, the entire process is done serially.
On the contrary, we minimize hardware idle time by utilizing software pipelines and distributing workload across CPU and GPU.
Research on heterogeneous computing is receiving attention in high performance computing.
Shee et al.~\cite{shee2008heterjpeg} conducted a case study on JPEG encoders on Application Specific Instruction-set Processors (ASIPs).
They evaluated two parallel programming patterns: master-slave and pipeline.
The master-slave model utilized task management and data-parallelism.
In the pipelined model, different ASIP processors were responsible for different stages of JPEG encoding.
L. Chen et al.~\cite{MapReduceCPUGPU2012} proposed a similar idea for MapReduce applications.
The authors developed two scheduling schemes, namely master-slave and pipeline model, on integrated CPU-GPU systems.
Data-parallelism and pipeline-parallelism were utilized separately.
The Qilin framework~\cite{QilinFramework} and CHC framework~\cite{CHCFramework} showed possibilities of a cooperative CPU-GPU computation of a CUDA application.
Both frameworks dynamically partitioned the workloads using their profiling based partitioning models.
Qilin used an empirical approach recording new execution to a database while CHC applied a heuristic approach.
The partitioning schemes were designed for CUDA applications and only supported data-parallelism.
The CAP scheduler~\cite{CAP2013} supports dynamic workload scheduling on CPU-GPU systems.
Profiling and workload partitioning are performed at run-time.
CAP profiles a small portion of the workload, verifies the accuracy of the ratio and then uses the ratio for the remaining of workload.
Although it can effectively partition workload, it only supports data-parallelism.
Our proposed partitioning scheme, in comparison, is designed specifically for JPEG decoder.
The workload is partitioned without user intervention, and the CPU and the GPU jointly perform the decoding tasks cooperating data-, task- and pipeline-parallelism.
\section{Conclusions}
\label{sec:conclusions}
We have introduced a novel JPEG decoding scheme for heterogeneous
architectures consisting of a CPU and an OpenCL-programmable GPU. Our
method employs an
offline profiling step to determine the performance of a system's CPU and GPU
with respect to JPEG decoding.
We apply multivariate polynomial regression analysis
to derive closed forms that characterize the performance of a given
CPU-GPU combination. Image entropy and the image dimensions are
the sole parameters for our performance model. At run-time, closed forms
are evaluated for a given image to estimate execution times and load-balance
the decoding workload between the CPU and the GPU.
Our
run-time partitioning scheme exploits task, data and pipeline
parallelism by scheduling the non-parallelizable entropy decoding task on the
CPU, whereas IDCT, color conversion and
upsampling are conducted on both the CPU and the GPU. Our kernels have been
optimized for GPU memory hierarchies.
We have implemented the proposed method in the context of the libjpeg-turbo
library, which is an industrial-strength JPEG encoding and decoding engine.
Irrespective of the GPU's computational power, our heterogeneous partitioning scheme
always achieves an improvement over the SIMD-version of libjpeg-turbo. The
results show speedups up to 8.5x over the sequential version and up to 4.2x
over the SIMD version of libjpeg-turbo. We have shown that our approach
achieves up to 95\% of the theoretically attainable speedup, with an average of 88\%. With
the availability of GPU accelerators on desktops and embedded devices such as
tablets and smartphones, heterogeneous JPEG image decompression will enhance
image viewing experiences ranging from personal photos to very large image
applications in medical imaging and astronomy.
\bibliographystyle{abbrvnat}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 6,978
|
'use strict';
/**
* @ngdoc overview
* @name deffileApp
* @description
* # deffileApp
*
* Main module of the application.
*/
angular
.module('deffileApp', [
// 'ngAnimate'
,'ngCookies'
,'ngResource'
,'ngRoute'
// ,'ngSanitize'
// ,'ngTouch'
])
.config(function ($routeProvider) {
$routeProvider
.when('/', {
templateUrl: '/admin/views/main.html',
controller: 'MainCtrl'
})
.when('/about', {
templateUrl: '/admin/views/about.html',
controller: 'AboutCtrl'
})
.otherwise({
redirectTo: '/'
});
});
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 782
|
// RobotBuilder Version: 2.0
//
// This file was generated by RobotBuilder. It contains sections of
// code that are automatically generated and assigned by robotbuilder.
// These sections will be updated in the future when you export to
// Java from RobotBuilder. Do not put any code or make any change in
// the blocks indicating autogenerated code or it will be lost on an
// update. Deleting the comments indicating the section will prevent
// it from being updated in the future.
package org.usfirst.frc5933.Caroline;
import org.usfirst.frc5933.Caroline.commands.*;
import edu.wpi.first.wpilibj.smartdashboard.SmartDashboard;
import edu.wpi.first.wpilibj.Joystick;
import edu.wpi.first.wpilibj.buttons.*;
/**
* This class is the glue that binds the controls on the physical operator
* interface to the commands and command groups that allow control of the robot.
*/
public class OI {
//// CREATING BUTTONS
// One type of button is a joystick button which is any button on a
//// joystick.
// You create one by telling it which joystick it's on and which button
// number it is.
// Joystick stick = new Joystick(port);
// Button button = new JoystickButton(stick, buttonNumber);
// There are a few additional built in buttons you can use. Additionally,
// by subclassing Button you can create custom triggers and bind those to
// commands the same as any other Button.
//// TRIGGERING COMMANDS WITH BUTTONS
// Once you have a button, it's trivial to bind it to a button in one of
// three ways:
// Start the command when the button is pressed and let it run the command
// until it is finished as determined by it's isFinished method.
// button.whenPressed(new ExampleCommand());
// Run the command while the button is being held down and interrupt it once
// the button is released.
// button.whileHeld(new ExampleCommand());
// Start the command when the button is released and let it run the command
// until it is finished as determined by it's isFinished method.
// button.whenReleased(new ExampleCommand());
// BEGIN AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=DECLARATIONS
public JoystickButton rBumper;
public JoystickButton xButton;
public JoystickButton aButton;
public JoystickButton yButton;
public JoystickButton bButton;
public JoystickButton backButton;
public JoystickButton lBumper;
public JoystickButton rToggle;
public JoystickButton lToggle;
public JoystickButton startButton;
public Joystick driverJoystick;
public JoystickButton xButtonSub;
public JoystickButton aButtonSub;
public JoystickButton bButtonSub;
public JoystickButton yButtonSub;
public JoystickButton rBumperSub;
public JoystickButton lBumperSub;
public JoystickButton rToggleSub;
public JoystickButton lToggleSub;
public JoystickButton backButtonSub;
public JoystickButton startButtonSub;
public Joystick subsystemJoystick;
// END AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=DECLARATIONS
public OI() {
// BEGIN AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=CONSTRUCTORS
subsystemJoystick = new Joystick(1);
//Button Object = new Object(Joystick Object, Button #)
//determine the button # by using the driver station
startButtonSub = new JoystickButton(subsystemJoystick, 8);
startButtonSub.whenPressed(new ToggleClimber());
backButtonSub = new JoystickButton(subsystemJoystick, 7);
backButtonSub.whenPressed(new /*ToggleClaw()*/NullCommand());
lToggleSub = new JoystickButton(subsystemJoystick, 9);
lToggleSub.whenPressed(new FlyFaster());
rToggleSub = new JoystickButton(subsystemJoystick, 10);
rToggleSub.whenPressed(new FlySlower());
lBumperSub = new JoystickButton(subsystemJoystick, 5);
lBumperSub.whenPressed(new ToggleHopper());
rBumperSub = new JoystickButton(subsystemJoystick, 6);
rBumperSub.whenPressed(new ToggleFlywheel());
yButtonSub = new JoystickButton(subsystemJoystick, 4);
yButtonSub.whenPressed(new /*ToggleShootingVision()*/NullCommand());
bButtonSub = new JoystickButton(subsystemJoystick, 2);
bButtonSub.whenPressed(new AgitateSlower());
aButtonSub = new JoystickButton(subsystemJoystick, 1);
aButtonSub.whenPressed(new AgitateFaster());
xButtonSub = new JoystickButton(subsystemJoystick, 3);
xButtonSub.whenPressed(new /*FlyWheelMode()*/ NullCommand());
driverJoystick = new Joystick(0);
startButton = new JoystickButton(driverJoystick, 8);
startButton.whileHeld(new StopFlaps());
lToggle = new JoystickButton(driverJoystick, 9);
lToggle.whenPressed(new /*DriveToPeg()*/NullCommand());
rToggle = new JoystickButton(driverJoystick, 10);
rToggle.whenPressed(new /*AlignToBoiler()*/NullCommand());
lBumper = new JoystickButton(driverJoystick, 5);
lBumper.whileHeld(new NullCommand());
backButton = new JoystickButton(driverJoystick, 7);
backButton.whenPressed(new NullCommand());
bButton = new JoystickButton(driverJoystick, 2);
bButton.whenPressed(new FrontFlapClose());
yButton = new JoystickButton(driverJoystick, 4);
yButton.whenPressed(new BackFlapClose());
aButton = new JoystickButton(driverJoystick, 1);
aButton.whenPressed(new FrontFlapOpen());
xButton = new JoystickButton(driverJoystick, 3);
xButton.whenPressed(new NullCommand());
rBumper = new JoystickButton(driverJoystick, 6);
rBumper.whileHeld(new NullCommand());
// SmartDashboard Buttons
//SmartDashboard is the console used to view feedback from the robot, from putData to putString and putNumber
SmartDashboard.putData("ComplexDriveStraight", new ComplexDriveStraight());
SmartDashboard.putData("RedAttackGearPosition1", new RedAttackGearPosition1());
SmartDashboard.putData("RedAttackGearPosition2", new RedAttackGearPosition2());
SmartDashboard.putData("RedAttackGearPosition3", new RedAttackGearPosition3());
SmartDashboard.putData("BlueAttackGearPosition1", new BlueAttackGearPosition1());
SmartDashboard.putData("BlueAttackGearPosition2", new BlueAttackGearPosition2());
SmartDashboard.putData("BlueAttackGearPosition3", new BlueAttackGearPosition3());
SmartDashboard.putData("SimpleDriveStraight", new SimpleDriveStraight());
// END AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=CONSTRUCTORS
}
// BEGIN AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=FUNCTIONS
//classic getters and setters.
public Joystick getDriverJoystick() {
return driverJoystick;
}
public Joystick getSubsystemJoystick() {
return subsystemJoystick;
}
// END AUTOGENERATED CODE, SOURCE=ROBOTBUILDER ID=FUNCTIONS
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,731
|
Q: How do I keep a user Authenticated in DNN 5.6.6 when using RadEditor's Image Manager? I have a RadEditor on a page, and it is breaking authentication on the next postback after I have closed the Image Manager Dialog.
I have tried adding this:
reThePolicy.DialogOpener.AdditionalQueryString = string.Format("&PortalID={0}", PortalId);
in the codebehind, and in IE9 it fails the first time, but every other time thereafter it works. In other browsers, it never works.
Any help is appreciated
A little background -- this is only happening on child portals, and only for users that do not exist in other portals.
in the ascx (the paths and whatnot are declared in the codebehind, but when the imagemanager pop-up doesn't kill authentication, that all works):
<telerik:RadEditor StripFormattingOptions="MSWordRemoveAll" ID="reThePolicy" runat="server"
Width="769px" Style="margin: 6px;">
<Content></Content>
</telerik:RadEditor>
in my web config:
<htmlEditor defaultProvider="RadEditorProvider">
<providers>
<clear />
<add name="FckHtmlEditorProvider" type="DotNetNuke.HtmlEditor.FckHtmlEditorProvider.FckHtmlEditorProvider, DotNetNuke.FckHtmlEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Fck/" CustomConfigurationPath="~/Providers/HtmlEditorProviders/Fck/custom/FCKConfig.js" EnhancedSecurityDefault="false" SecureConfigurationPath="~/Providers/HtmlEditorProviders/Fck/custom/FCKConfigSecure.js" ImageGalleryPath="~/Providers/HtmlEditorProviders/Fck/fckimagegallery.aspx" ImageUploadPath="~/Providers/HtmlEditorProviders/Fck/fckimagegallery.aspx" ImageAllowedFileTypes="gif,png,bmp,jpg" FlashGalleryPath="~/Providers/HtmlEditorProviders/Fck/fckimagegallery.aspx" FlashUploadPath="~/Providers/HtmlEditorProviders/Fck/fckimagegallery.aspx" FlashAllowedFileTypes="fla,swf" LinksGalleryPath="~/Providers/HtmlEditorProviders/Fck/fcklinkgallery.aspx" DynamicStylesGeneratorPath="~/Providers/HtmlEditorProviders/Fck/FCKStyles.aspx" DynamicStylesCaseSensitive="true" DynamicStylesGeneratorFilter="controlpanel|filemanager|mainmenu|wizard" StaticStylesFile="~/Providers/HtmlEditorProviders/Fck/FCKeditor/fckstyles.xml" StylesDefaultMode="Static" DynamicCSSGeneratorPath="~/Providers/HtmlEditorProviders/Fck/FCKCSS.aspx" StaticCSSFile="~/Providers/HtmlEditorProviders/Fck/FCKeditor/editor/css/fck_editorarea.css" CSSDefaultMode="static" spellCheck="ieSpell" AvailableToolbarSkins="Office2003,Silver" DefaultToolbarSkin="Office2003" AvailableToolBarSets="DNNDefault,Default,NoGallery,Basic" DefaultToolbarSet="DNNDefault" DefaultImageGallerySkin="Default" DefaultFlashGallerySkin="Default" DefaultLinksGallerySkin="Default" FCKDebugMode="false" UseFCKSource="false" OptionsOpenMode="ShowModalDialog" CustomOptionsDialog="Admin" />
<add name="RadEditorProvider" type="Telerik.DNN.Providers.RadEditorProvider" providerPath="~/DesktopModules/TelerikWebUI" AutoCreatePaths="true" />
<!--Upgraded by DotNetNuke.TelerikEditorProvider version 5.4.2 - Date: 11/8/2010 4:07:51 PM-->
<!--<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />-->
<!--Upgraded by DotNetNuke.TelerikEditorProvider version 5.4.2 - Date: 11/20/2010 12:52:19 AM-->
<!--<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />-->
<!--Upgraded by DotNetNuke.TelerikEditorProvider version 5.6.0 - Date: 11/20/2010 12:52:20 AM-->
<!--<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />-->
<!--Upgraded by DotNetNuke.TelerikEditorProvider version 5.6.3 - Date: 9/14/2011 9:51:59 AM-->
<!--<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />-->
<!--Upgraded by DotNetNuke.TelerikEditorProvider version 5.6.3 - Date: 1/19/2012 12:44:58 PM-->
<!--<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />-->
<add name="TelerikEditorProvider" type="DotNetNuke.HtmlEditor.TelerikEditorProvider.EditorProvider, DotNetNuke.HtmlEditor.TelerikEditorProvider" providerPath="~/Providers/HtmlEditorProviders/Telerik/" toolsFile="~/Providers/HtmlEditorProviders/Telerik/Config/ToolsDefault.xml" configFile="~/Providers/HtmlEditorProviders/Telerik/Config/ConfigDefault.xml" FilterHostExtensions="True" />
</providers>
</htmlEditor>
A: So I posted this on a number of Forums, and I got the answer from Telerik.
Their suggestion, although humorous [Don't use our product :D ]
was the one that worked. It is:
Our suggestion is to not use the -telerik:RadEditor- in the ascx file as a standalone control in DNN, but to use the integrated in DNN -dnn:TextEditor- control.
Please, also note that DotNetNuke.TelerikEditorProvider provider of RadEditor is developed and supported by DNN Corp only.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 6,073
|
Get a Virtual Phone Number for Minot, North Dakota
Minot, your Ninja is waiting.
Minot, North Dakota - Facts & Figures
Minot is a city in and the county seat of Ward County, North Dakota, in the state's north-central region. It is most widely known for the Air Force base located approximately 15 miles north of the city. With a population of 40,888 at the 2010 census, Minot is the fourth largest city in the state and a trading center for a large portion of northern North Dakota, southwestern Manitoba, and southeastern Saskatchewan. Founded in 1886 during the construction of the Great Northern Railway, Minot is also known as "Magic City", commemorating its remarkable growth in size over a short time. Minot is the principal city of the Minot Micropolitan Statistical Area, a micropolitan area that covers McHenry, Renville, and Ward counties and had a combined population of 69,540 at the 2010 census. In 2012, it was estimated that the population of the Minot Micropolitan Area was 73,146. Minot came into existence in 1886, when James J. Hill's Great Northern Railway ended its push through the state for the winter, after having difficulty constructing a trestle across Gassman Coulee. It was the end of the railway's line, so whenever a train came into the town and the stop was announced, the conductor would call out "Minot, this is Minot, North Dakota, prepare to meet your doom." A tent town sprang up overnight, as if by magic, and thus the city came to be known as the Magic City; in the next five months, the population increased to over 5,000 residents, further adding to the nickname's validity.
http://www.minotnd.org/
http://www.visitminot.org/
http://www.minotchamber.org/
Industries: Agriculture, Energy, Unmanned Aerial Systems, Distribution
ZIP Codes: 58701-58705, 58707
Surrounding Cities: Carnduff, CAN , Melita, CAN , Deloraine, CAN , Oxbox, CAN , Boissevain, CAN
Here are some more cities in North-dakota where we offer service!
Bismarck701
Fargo701
Grand Forks701
Minot, get your Ninja working for you right away.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,818
|
Le Synaspismós, dont le nom entier en grec , Synaspismós tīs Aristerás tōn Kinīmátōn kai tīs Oikologías, signifie Coalition de la gauche, des mouvements et de l'écologie, et s'abrège couramment en ΣΥΝ - SYN) est un ancien parti politique grec, qui a existé de 1991 à 2013. Il est devenu en 2004 membre de la Coalition de la gauche radicale (SYRIZA), puis s'est dissous en son sein lors de la transformation de celle-ci en parti politique.
Histoire du parti
Le SYN est l'héritier direct de la coalition formée en vue des législatives de 1989 entre le Parti communiste de Grèce (KKE, prosoviétique) et sa branche dissidente dite , favorable à l'intégration de la Grèce à l'Union européenne. Après la chute de l'Union soviétique, la gauche grecque n'est pas épargnée par la crise profonde qui frappe les formations d'obédience marxiste-léniniste dans le monde entier et l'aile dite du parti prend la décision d'exclure les du KKE , qui constituaient une part importante des militants et auxquels appartenaient près de 45 % des membres du comité central, dont le secrétaire général du parti de l'époque, . Les exclus forment en 1991 un parti désormais totalement indépendant du Parti communiste et reprennent le nom de Synaspismós.
En 1993, le SYN remporte 2,94 % des voix aux législatives, score insuffisant pour obtenir un siège au Parlement. Cet échec a pour effet immédiat la démission de la présidente du parti, María Damanáki.
Un an plus tard, en 1994, le SYN, présidé par Níkos Konstantópoulos, parvient néanmoins à rassembler 6,26 % des suffrages aux élections européennes, ce qui constitue son meilleur score à ce jour.
Aux législatives de 1996, le SYN améliore son score de 1993 en obtenant 5,1 % des voix.
Le parti traverse une crise grave après les législatives de 2000, quand un petit groupe de cadres emmenés par Nikos Bistis se prononcent en faveur d'un rapprochement stratégique avec les sociaux-démocrates du PASOK. Accusant le parti de , Bistis et ses partisans font sécession et fondent une nouvelle formation politique (le Mouvement de réforme et de modernisation de la gauche, abrégé en AEKA), dissoute très rapidement, au moment de l'entrée de son président dans le gouvernement de Konstantínos Simítis.
C'est en reprenant l'expression de que l'ex-présidente du parti, María Damanáki, rompt avec le SYN en 2003 pour rejoindre les rangs du PASOK.
Aux législatives de 2004, le SYN s'allie avec plusieurs petits partis d'extrême-gauche (trotskystes et maoïstes entre autres) ainsi que du petit Parti écologiste-communiste (AKOA) dont il est proche depuis 2000 dans le cadre de la SYRIZA. Le Synaspismós obtient six députés ce qui crée des tensions dans la coalition, qu'il finit par quitter.
Aux élections européennes de 2004, ces alliances sont néanmoins rompues en raison d'une nouvelle crise interne, résolue par l'élection d'un nouveau président, Alékos Alavános. En , ce dernier annonce qu'il quitte son poste mais conserve la direction de la SYRIZA. Il est remplacé en par Aléxis Tsípras.
Face à la crise économique, Synaspismos s'engage dans l'action sociale, orientant en particulier les efforts de ses adhérents vers la constitution de banques solidaires de médicaments.
Afin de pouvoir bénéficier de la prime majoritaire ( en plus de ceux élus à la proportionnelle intégrale) accordée aux partis politiques, et non pas aux coalitions, la nécessité de former un parti unitaire apparaît alors. Le , la nouvelle SYRIZA décide d'une nouvelle organisation en fusionnant ses différentes composantes. Le Synaspismós est alors dissous.
Résultats électoraux
Élections parlementaires
Au sein de SYRIZA.
Élections européennes
Le parti est dans SYRIZA.
Notes et références
Voir aussi
Bibliographie
Liens externes
Parti politique fondé en 1991
Parti politique disparu en 2013
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 8,500
|
Laminectomy is used as a last resort when all non-surgical techniques do not relieve the pressure on the spinal canal. It may be done along with other spinal surgeries like diskectomy, spinal fusion and formainotomy.
A hospital stay for 1 to 2 may be required. There will be some pain in the back for which you will be given medications.
Activities that cause stress on the back must be avoided for a few weeks. The sutures/staples will be removed after about 2 weeks.
Physiotherapy will be recommended to avoid formation of blood clots in the legs and improving flexibility.
If spinal decompression is performed along with any other procedure, then the recovery time may be upto 6 months.
The patient may have reaction to anesthetic medications, breathing problems, infection and excessive bleeding during the surgery.
The spinal nerve may be injured resulting in pain and numbness.
Spinal Decompression will relieve the symptoms but will not prevent spinal problems from occurring in future. The symptoms may re-appear after a few years.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 3,455
|
\section{Introduction}
Visual object tracking has attracted public attention due to
its extensive applications in the field of computer vision,
such as virtual reality, medical image classification and activity analysis.
In general, most of existing methods can be categorized into two branches, i.e., tracking-by-detection and tracking-by-segmentation.
Tracking-by-detection methods~\cite{MEEM2014,Babenko2011Robust,Grabner2008Semi,6619157,Danelljan2016ECO,Hong2015MUlti,Danelljan2014Accurate,Jiang2019Robust,HuangXLY2019,Feng2020Tracking} aim to estimate the locations of the target object in the video sequences by using a rectangle bounding box around the target object,
as shown in Figure~\ref{img::comparision}(a).
In contrast, tracking-by-segmentation methods~\cite{Yeo2017Superpixel,Tracking-by-Segmentation2018} pixel-wisely segment the object from background and then track the target according to the segmentation results, as shown in Figure~\ref{img::comparision}(b).
\begin{figure}[htb]
\label{img::comparision}
\subfigure[]{
\begin{minipage}[t]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{dec.eps}
\end{minipage}
}
\subfigure[]{
\begin{minipage}[t]{0.5\textwidth}
\centering
\includegraphics[width=\textwidth]{seg.eps}
\end{minipage}
}
\caption{Comparison of tracking-by-segmentation and tracking-by-detection. In a video sequence, as shown in (a), the detection-based tracking generally uses a bounding box to locate the target, while (b) illustrates that segmentation-based tracking segments the target and then locates the target according to the segmentation mask.}
\end{figure}
Early tracking-by-segmentation works usually adopt pixel-wise segmentation strategies~\cite{5539810,Godec2011Hough,Duffner2014PixelTrack}.
For example, Chad et al.~\cite{5539810} propose a probability framework based on the joint Gaussian distribution over all the pixels to jointly solve the segmentation and tracking. Godec et al.~\cite{Godec2011Hough} use the generalized Hough transform and online learning technology to roughly separate the object and background pixel by pixel. Duffner et al.~\cite{Duffner2014PixelTrack} integrate generalized Hough transform with probabilistic segmentation approach together for pixel-based object tracking.
Although these segmentation methods can distinguish the foreground and background, their abilities to handle the heavily occlusion and clutter are limited because they fail to consider the internal structural information of the target object.
Besides, pixel-level processing usually has high computational complexity.
To alleviate these issues, some works utilize super-pixel methods to model non-rigid and deformable object tracking and reduce dimensions. Yang et al.~\cite{yang2014robust} propose a model for appearance discrimination, which extract middle-level features from hyper-pixels to distinguish target objects and background. Wang et al.~\cite{7875137} propose a constraint graph labeling algorithm, which take into account the internal structure information of the graph. The nodes of the graph model represent hyper-pixels, and the space, time and appearance fitness constraints of the bottom layer are encoded by the edge. Yeo et al.~\cite{Yeo2017Superpixel} propose to use Absorbing Markov Chain (AMC) model for super-pixel segmentation.
However, one main limitation of the above approaches is that they usually individually conducts super-pixel feature representation and learning (or labeling) which may lead to sub-optimal results.
To overcome above issues, in this paper, we propose to utilize Graph Convolutional Network (GCN) model for super-pixel based object segmentation and tracking.
The proposed model provides a general framework that integrates the super-pixel feature representation and labeling together in an end-to-end framework.
The main advantage of the proposed GCN method is that it provides an effective context-aware representation method for each hyper-pixel by using the spatial-temporal structural information of different hyper-pixels.
Also, we provide an efficient way to implement the proposed model.
Overall, the main contributions of this paper are summarized as follows.
\begin{itemize}
\item We propose to employ a graph convolutional network model to obtain context-aware feature representations for super-pixels which improves tracking performance.
\item We present to design an efficient and effective optimization algorithm to solve the proposed model.
\item Experiments on several widely used benchmark data-sets demonstrate the superiority of our model over current tracking approaches.
\end{itemize}
\section{Related Work}
\subsection{Graph convolutional network}
Graph convolutional networks(GCNs) have been commonly studied in computer vision field.
As a branch of GNNs, GCNs can extend the convolutional architecture to arbitrary graph-structural data including both regular and irregular structures~\cite{DuvenaudConvolutional,KipfSemi,NiepertLearning}. For example, M-GLCN~\cite{M-GLCN2019} is proposed for image co-saliency estimation task.~\cite{Multi-Label2019} demonstrates the advantages of GCNs on multi-label classification task. ~\cite{Qiu2018DeepInf} applies GCN to social influence prediction. PH-GCN~\cite{PH-GCN2019} is designed for person Re-ID problem.
Park et al, ~\cite{ParkSymmetric} propose GALA for unsupervised graph representation and learning.
Jiang et al, ~\cite{GLMNet2019} propose GLMNet for feature matching problem by incorporating both smoothing and resharpening together~\cite{ParkSymmetric}.
Different from these works~\cite{GLMNet2019,ParkSymmetric}, we adapt both smoothing and resharpening GCN for visual object segmentation and tracking problem. Also, we develop a simple and effective graph network, followed by an efficient optimization algorithm, for our tracking task based on SGCN~\cite{WuSimplifying}.
Recently, GCN is also applied to tracking-by-detection problems. Tu et al.~\cite{tu2019visual} propose to use GCNs to convert the heterogeneous features extracted from CNN into structured information. Gao et al.~\cite{GCT} integrate two different GCNs into existing Siamese methods~\cite{2016Fully} to model the appearance of an object fully considering the space-time information of the object context. However, the detection-based tracking methods cannot display the segmentation mask of the tracking target. In contrast, in this paper, we mainly focus on integrating GCN into the segmentation-based tracking framework.
\subsection{Tracking-by-segmentation}
Here, we briefly review some related works on segmentation based tracking problem~\cite{Godec2011Hough,OGBDT2015,Yeo2017Superpixel,Tracking-by-Segmentation2018}. In work~\cite{Godec2011Hough}, it extends Hough Forests to online mode and couples the rough GrabCut segmentation with voting-based detection and back-projection. Son et al.~\cite{OGBDT2015} propose an online tracking algorithm based on online gradient enhancement decision tree. Yeo et al.~\cite{Yeo2017Superpixel} present a tracking-by-segmentation framework using absorbing markov model to better distinguish the foreground and background. Lee et al.~\cite{Tracking-by-Segmentation2018} design an extended version of~\cite{Yeo2017Superpixel} by replacing the simple feature representation in ~\cite{Yeo2017Superpixel} with multi-level convolutional features. Different from previous works, this paper propose a contextual feature representations with stronger discrimination for super-pixels by using graph convolution representation model. In addition, we conduct feature learning and labeling together in an end-to-end network framework, which helps to accurately segment the target of each frame.
\section{The proposed Model}
Given a candidate region of the target object,
we first partition it into non-overlapping superpixels via Simple Linear Iterative Clustering (SLIC) algorithm~\cite{Achanta2012SLIC} and then
extract the feature descriptor for each superpixel.
Then, we aim to assign a weight value to represent its possibility of belonging to the target. To this end, we formulate this task as graph representation and labeling and propose
to employ a graph convolutional network model (GCN) for this task. The whole flow of the proposed network is shown in Figure~\ref{img::framework}.
\begin{figure*}[htb]
\label{img::framework}
\centering
\includegraphics[scale=0.45]{Frameall.eps}
\caption{Architecture of the proposed GCN network for tracking-by-segmentation.}
\label{img::framework}
\end{figure*}
\subsection{Graph construction}
Let $X=(X^{t-1}, X^{t})$ be the
collection of feature descriptors of super-pixels of two consecutive frames t-1, t where $X^{t-1}\in \mathbb{R}^{d\times n_{t-1}}$ and $X^{t}\in \mathbb{R}^{d\times n_{t}}$ and let $ n_{t} $ denote the number of super-pixels in frame t.
We first construct a spatial-temporal graph as $G = (V,E)$, where $V = \{V^{t-1},V^{t}\}$ denotes the superpixels node set of two consecutive frames and edge set $E$ represents the relationships among different superpixels. There are two kinds of edges in $G = (V,E)$, which represent the similarities among nodes within each frame and between different continuous frames.
First, the weight of edge in each frame is calculated as
\begin{equation}
\begin{aligned}
\label{eq::eq1}
A_{ij}^{t} = \exp\Big(-\dfrac{\left\|x_{i}^{t}-x_{j}^{t}\right\|_2}{\sigma}\Big),
\end{aligned}
\end{equation}
where $ x_{i}^{t}, x_{j}^{t} $ denote the feature descriptors of any two node pairs in current frame $t$ and $ \sigma $ is a scaling parameter. Similarly, we can also obtain $A^{t-1}$ for the previous frame.
Then, for the edges connected between different consecutive frames $B \in \mathbb{R}^{n_{t-1}\times n_t}$, we also use EPPM~\cite{Bao2014Fast} to extract the optical flow information at the pixel level, as suggested in work~\cite{Yeo2017Superpixel}.
Finally, we obtain the spatial-temporal graph $A\in \mathbb{R}^{n\times n}, n=n_{t-1}+n_t$ of the two consecutive frames as
\begin{equation}
\label{eq::AA}
A=\left(\begin{array}{ll}{A^{t-1}} & {B} \\ {B^{T}} & {A^{t}}\end{array}\right)
\end{equation}
The constructed graph $G(V,E)$ is a symmetric graph.
\subsection{Graph convolutional network}
Recently, graph convolutional network (GCN) has been commonly employed in many computer vision tasks, such as image co-saliency estimation~\cite{M-GLCN2019}, social influence prediction~\cite{Qiu2018DeepInf} and multi-label classification ~\cite{Multi-Label2019}, etc. In this paper, we adapt it for our tracking problem. Specifically, we formulate the problem of superpixel labeling as node labeling on the constructed superpixel graph $G(V,E)$ by employing a GCN model. In general, GCN consists of an input layer, several propagation layers and a final output layer.
Given superpixel features $X\in \mathbb{R}^{n\times d}$ and graph adjacency matrix $A\in \mathbb{R}^{n\times n}$ Eq.(~\ref{eq::AA}), then we can use a two-layer GCN model proposed in work~\cite{KipfSemi} as
\begin{equation}
\begin{aligned}
\label{eq::GCN}
z=&\phi(X, A;W^{(0)},W^{(1)}) \\
=& \operatorname{softmax}\big(\hat{A} \operatorname{ReLU}\big(\hat{A} X W^{(0)}\big) W^{(1)} + \textbf{1}b\big)
\end{aligned}
\end{equation}
where $\hat{A}=D^{-\frac{1}{2}} A D^{-\frac{1}{2}}$ and $\textbf{1}=(1,1\cdots 1)$. Parameter $ W^{(0)}\in \mathbb{R}^{d \times h} $ represents the weight matrix of the first layer and $ W^{(1)}\in \mathbb{R}^{h \times 1}$ represents the weight matrix of the second output layer. Here, $h$ denotes the number of hidden layer units. $b$ is a bias term.
The final output $z\in \mathbb{R}^{n \times 1}$ denotes the final label prediction for all graph nodes, i.e., $z_i$ represents the confidence/weight that the $i$-th superpixel belongs to the target object.
To maintain more discriminative information, Park et al.,~\cite{ParkSymmetric} propose GALA (Graph convolutional Autoencoder using Laplacian smoothing and sharpening) for graph node feature representation.
Using GALA~\cite{ParkSymmetric}, we
propose a two-layer graph network for superpixel labeling as
\begin{equation}
\begin{aligned}
\label{eq::GCN}
z=&\phi(X, A;W^{(0)},W^{(1)}) \\% Z=f(X, A)=
= & \operatorname{softmax}\big(\hat{A}_m \operatorname{ReLU}\big(\hat{A}_h X W^{(0)}\big) W^{(1)} + \textbf{1}b \big)
\end{aligned}
\end{equation}
Here, $\hat{A}_m$ and $\hat{A}_h$ are defined respectively as~\cite{ParkSymmetric}
\begin{equation}
\begin{aligned}
\label{eq::mode1_our}
\hat{A}_m & = I - \lambda_{1}(I-D^{-\frac{1}{2}} A D^{-\frac{1}{2}}) \\
\hat{A}_h & =I + \lambda_{2}(I-D^{-\frac{1}{2}} A D^{-\frac{1}{2}})
\end{aligned}
\end{equation}
where $ D $ denotes a diagonal matrix with $ D_{ii} = \sum_{j} A_{ij} $ and $\lambda_1, \lambda_2$ are two parameters.
In addition, to further simplify the nonlinear mapping between our GCN layers, as suggested in work~\cite{WuSimplifying}, we propose to transform the nonlinear the above nonlinear model into a simple linear model.
Formally, by removing nonlinear function $\mathrm{ReLU}$, the above model becomes,
\begin{equation}
\begin{aligned}
\label{eq::GCN1}
z& =\phi \big(X, A;W^{(0)},W^{(1)}\big)\\
& =\operatorname{softmax}\big(\hat{A}_m\hat{A}_h X W^{(0)}W^{(1)} + \textbf{1}b \big)
\end{aligned}
\end{equation}
It can be rewritten compactly as
\begin{equation}
\begin{aligned}
\label{eq::GCN2}
z=\phi(X, A;W)=\operatorname{softmax}\big(\hat{A}_m\hat{A}_h X W + \textbf{1}b \big)
\end{aligned}
\end{equation}
where $W = W^{(0)}W^{(1)}\in \mathbb{R}^{d\times 1}$.
In the following, we propose an effective way to train the proposed model~Eq.(\ref{eq::GCN2}). The optimization algorithm of loss function is summarized in Algorithm~\ref{alg:algorithm}.
\subsection{Loss function}
In this paper, we propose to learn the optimal network parameters $\{W, b\}$ by employing a semi-supervised learning manner.
Specifically, the already target tracked result in previous $t-1$ frame provides the labelled information to guide the training of the proposed model.
Formally, let $f=(f_1, f_2,\cdots f_{n_{t-1}})\in \mathbb{R}^{n_{t-1}\times 1}$ be the indicative vector of target tracked/segemented result in the previous frame, i.e.,
$f_i=1$ if the $i$-th superpixel belongs to target in previous $t-1$ frame, and $f_i=0$ otherwise.
Then, we adopt the flexible manifold ranking model~\cite{wang2018flexible} to define our loss function as
\begin{equation}
\begin{aligned}
\label{eq::Fusion Model}
& \mathcal{L}(y,W,b) = \\
& \|\phi(X, A;W,b) -y\|^2_2
+ \alpha y^TL_Sy + \beta \|(y^{t-1}-f)\|_2^2
\end{aligned}
\end{equation}
where $ y=(y^{t-1},y^t) $ is a variable vector to be optimized and
$ L_S = D-A $ is the Laplacian matrix.
The first residual term denotes the label prediction loss.
The second term is the smoothing constraint which encourages neighboring nodes have similar labels.
The last term denotes the label fitting term.
The main benefit of introducing the variable $y$ is to make the model be more flexible, as suggested in work~\cite{wang2018flexible}.
\section{Optimization}
In this section, we propose an effective algorithm to optimize the proposed model~Eq.(\ref{eq::GCN2}).
For efficiency consideration,
we propose to optimize the loss~Eq.(\ref{eq::Fusion Model}) by ignoring the softmax function.
In this case, we can derive a simple update algorithm to optimize the loss function.
Model~Eq.(\ref{eq::GCN2}) contains three trainable parameters, i.e., network work parameters $\{W, b\}$ and variable $y$. They are optimized by alternatively optimize them until convergence, as summarized in Algorithm~\ref{alg:algorithm}.
\textbf{(1) $\{W, b\}$-problem: Fix $ y $, update $\{W, b\}$.}
The problem becomes
\begin{equation}
\begin{aligned}
\label{eq::solve Model_1}
& \min_{W,b} \|\phi(X, A;W,b) -y\|^2_
\end{aligned}
\end{equation}
By ignoring the softmax function, it becomes
\begin{equation}
\begin{aligned}
\label{eq::solve Model_1}
\min_{W,b} \|(\hat{A}_m\hat{A}_h X)^{T}W +\textbf{1} b -y\|^2_2
\end{aligned}
\end{equation}
It is known that, the optimal $ W,b $ can be acquired by setting the first order derivative of Eq.(\ref{eq::solve Model_1}) with respect to variables $ W,b $ to zeros respectively which is given as
\begin{equation}
\begin{aligned}
\label{eq::solve Model_11}
&W^{*}=\left((\hat{A}_m\hat{A}_h X)^{T}\right)^{-1}(X(\textbf{1} b-y)
\end{aligned}
\end{equation}
\begin{equation}
\begin{aligned}
\label{eq::solve Model_12}
&b^* =\frac{1}{n} \textbf{1}^T \big(y-(\hat{A}_m\hat{A}_h X)^{T}W\big)
\end{aligned}
\end{equation}
\textbf{(2) y-problem: Fix $\{W, b\}$, update $ y $.} The problem becomes
\begin{equation}
\begin{aligned}
\label{eq::solve Model_2}
& \min_{y} \|(\hat{A}_m\hat{A}_h X)^{T}W +\textbf{1} b -y\|^2_2 + \alpha y^TL_Sy + \beta \|y^{t-1}-f\|_2^2 \\
& s.t.\ \ y_i\geq 0
\end{aligned}
\end{equation}
which is simply rewritten as
\begin{equation}
\begin{aligned}
\label{eq::solve Model_21}
\min_{y} \|y - q\|_2^2 +\alpha y^TL_Sy \ \ \ \ s.t.\ \ y_i \geq 0
\end{aligned}
\end{equation}
where $q=\Gamma ((\hat{A}_m\hat{A}_h X)^{T}W +1 b +\beta f) $ and $ \Gamma = diag(u^{t-1}, u^{t}) $,
$ u^{t-1} $ = $[1, 1, ..., 1] \in \mathbb{R}^{n^{t-1} \times 1} $, $ u^{t} $ = $ [0, 0, ..., 0] \in \mathbb{R}^{n^{t} \times 1} $.
We can simply prove that Eq.(~\ref{eq::solve Model_21}) can also be written as
\begin{equation}
\begin{aligned}
\label{eq::solve Model_42}
\min_{y} \|y - \hat{q}\|_2^2 +\alpha y^TL_Sy
\end{aligned}
\end{equation}
\noindent where $ {\hat{q}}_i = \max\{q_i,0\} $.
It is known that the Eq.(~\ref{eq::solve Model_42}) is a convex function, so its optimal solution must be non-negative. Thus, by setting the first derivative of Eq.(~\ref{eq::solve Model_42}) with respect to variable $ y $ to zero. Then the optimal $ y $ can be obtained as
\begin{equation}
\begin{aligned}
\label{eq::solve Model_43}
y^* = (I +\alpha L_S)^{-1}\hat{q}
\end{aligned}
\end{equation}
\begin{algorithm}[ht]
\caption{Optimization algorithm of~(Eq.(\ref{eq::GCN2}))}
\label{alg:algorithm}
\textbf{Input}: The feature descriptor $X=(X^{t-1}, X^{t})$, $X \in \mathbb{R}^{d\times n}$, indicative vector $f \in \mathbb{R}^{n\times 1}$ and two kinds of graph $ \hat{A}_m$, $\hat{A}_h $. \\
\textbf{Parameter}: $ \lambda_{1}=0.01 $, $\lambda_{2}=0.07$, $\alpha=0.001$, $ \beta=50 $, $ minError=1e-4 $,
$ maxLter=le-5 $. \\
\textbf{Output}: Variable vector. $y\in\mathbb{R}^{n \times 1}$. \\
\textbf{Initialize}: Indicator vector $f$ is initialized with the segmentation result of the previous frame, and linear regression parameter $W,b$ are set to $\textbf{0}$ and $0$.
\begin{algorithmic}[1]
\WHILE{not converged}
\STATE Update $ W $ by solving the problem Eq.(~\ref{eq::solve Model_11});
\STATE Update $ b $ by solving the problem Eq.(~\ref{eq::solve Model_12});
\STATE Update $ y $ by solving the problem Eq.(~\ref{eq::solve Model_43});
\ENDWHILE
\STATE \textbf{until} converge
\STATE \textbf{return} $ y $;
\end{algorithmic}
\end{algorithm}
\section{Experiment}
To evaluate the effectiveness of the proposed GCN-based tracking approach. We
test it on several datasets and compare it with some other related approaches.
This section provides both quantitative and qualitative results of our proposed approach.
\subsection{Datasets}
We evaluate the proposed algorithm on five benchmark datasets
including DAVIS~\cite{Perazzi2016A}, GBS~\cite{6126494,Lim2014Generalized,Lim2013Joint}, ST2~\cite{Li2013Video}, NR~\cite{OGBDT2015} and VS~\cite{Fukuchi2009Saliency} datasets.
The detail of these datasets are introduced below.
\begin{itemize}
\item DAVIS~\cite{Perazzi2016A} contains 50 different video sequences which covers the common challenges such as deformation, occlusion, and appearance changes in target segmentation and tracking tasks.
\item GBS (Generalized Background Subtraction)~\cite{6126494,Lim2014Generalized,Lim2013Joint} dataset contains 15 video sequences which
also includes deformation, occlusion, and low resolution.
\item ST2 (SegTrack v2)~\cite{Li2013Video} dataset contains 24 targets which are mainly for problems similar to GBS and some of them require tracking multiple ground truth objects.
\item NR (Non-rigid object tracking)~\cite{OGBDT2015} contains 10 challenging video sequences which are challenging mainly due to articulated and non-rigid deformable objects.
\item VS (Video Saliency)~\cite{Fukuchi2009Saliency} dataset consists of 10 sequences with complex backgrounds or multiple objects which considers only the most significant objects and ignores those that are too small to segment.
\end{itemize}
We compare the proposed method with six state-of-the-art tracking-by-segmentation methods and four tracking-by-detection methods, including AMCT~\cite{Yeo2017Superpixel}, OGBDT~\cite{OGBDT2015}, HT~\cite{Godec2011Hough}, SPT~\cite{Shu2011Superpixel}, PT~\cite{Duffner2014PixelTrack}, MUSTer~\cite{Hong2015MUlti}, DSST~\cite{Danelljan2014Accurate} and MEEM~\cite{MEEM2014}.
\subsection{Experimental Setup and Evaluation Metrics}
\begin{table*}
\small
\centering
\caption{Average overlap ratio of segmentation masks for tracking-by-segmentation algorithms.}
\label{table::table1}
\setlength{\tabcolsep}{4.5mm}{
\begin{tabular}{l | c| c | c | c | c | c c }
\hline
\hline
& Ours & AMCT~\cite{Yeo2017Superpixel} & OGBDT~\cite{OGBDT2015} & HT~\cite{Godec2011Hough} &SPT~\cite{Shu2011Superpixel} & PT~\cite{Duffner2014PixelTrack} \\
\hline
{DAVIS~\cite{Perazzi2016A}} & {\bf 62.0} & 59.2 & 44.9 & 33.1 & 27.1 & 26.1\\
{GBS~\cite{6126494,Lim2014Generalized,Lim2013Joint}} & {\bf 75.4} & 74.8 & 59.7 & 40.4 & 45.9 & 35.3\\
{ST2~\cite{Li2013Video}} & {\bf 64.0} & 58.8 & 47.6 & 43.0 & 26.3 & 21.2\\
{NR~\cite{OGBDT2015}} & {\bf 60.3} & 58.6 & 53.3 & 41.1 & 29.7 & 28.3\\
{VS~\cite{Fukuchi2009Saliency}} & {\bf 87.6} & 84.1 & 79.8 & 51.2 & 61.0 & 73.9\\
\hline
\hline
\end{tabular}}
\end{table*}
\begin{table*}
\small
\centering
\caption{Average overlap ratio of bounding boxes for tracking algorithms.}
\label{table::table2}
\vspace{0.1 ex}
\setlength{\tabcolsep}{1mm}{
\begin{tabular}{l | c |c |c |c| c| c | c| c| c c}
\hline
\hline
& \multicolumn{6}{c|}{tracking-by-segmentation algorithms}& \multicolumn{4}{c}{tracking-by-detection algorithms} \\
\cline{2-11}
~ & Ours & AMCT~\cite{Yeo2017Superpixel} & OGBDT~\cite{OGBDT2015} & HT~\cite{Godec2011Hough} & SPT~\cite{Shu2011Superpixel} & PT~\cite{Duffner2014PixelTrack} & MUSTer~\cite{Hong2015MUlti} & DSST~\cite{Danelljan2014Accurate} & MEEM~\cite{MEEM2014} \\
\hline
{DAVIS~\cite{Perazzi2016A}} &{\bf 61.8} & 60.9 & 50.0 & 35.8 & 43.2 & 41.6 & 25.9 & 58.4 & 52.7\\
{GBS~\cite{6126494,Lim2014Generalized,Lim2013Joint}} &{\bf 81.2} & 80.0 & 61.2 & 43.0 & 55.2 & 44.7 & 59.4 & 62.9 & 52.6\\
{ST2~\cite{Li2013Video}} &{\bf 68.6} & 64.8 & 50.2 & 44.9 & 53.5 & 32.2 & 58.8 & 62.0 & 59.5\\
{NR~\cite{OGBDT2015}} &{\bf 67.5} & 66.9 & 60.8 & 40.9 & 35.7 & 16.1 & 36.2 & 35.4 & 33.1\\
{VS~\cite{Fukuchi2009Saliency}} &{\bf 90.6} & 88.2 & 78.8 & 57.6 & 61.5 & 51.9 & 64.1 & 66.9 & 60.3\\
\hline
\hline
\end{tabular}}
\end{table*}
\begin{figure*}
\subfigure[]{
\begin{minipage}[t]{0.33\textwidth}
\centering
\includegraphics[width=\textwidth]{222.eps}
\end{minipage}
}
\subfigure[]{
\begin{minipage}[t]{0.33\textwidth}
\centering
\includegraphics[width=\textwidth]{111.eps}
\end{minipage}
}
\subfigure[]{
\begin{minipage}[t]{0.33\textwidth}
\centering
\includegraphics[width=\textwidth]{333.eps}
\label{img:precision}
\end{minipage}
}
\caption{Success and precision plots on five datasets: (a) success plots in terms of bounding box overlap ratio (b) success
plots in terms of segmentation overlap ratio (c) precision plots.}
\label{img::plots}
\end{figure*}
\begin{figure*}
\centering
\includegraphics[scale=0.65]{visualall.eps}
\caption{With challenges of non-rigid object deformation, occlusion, appearance changes and other challenges. From top to bottoms, we show segmentation and tracking results of \emph{fox1} in VS, \emph{tennis} in GBS, \emph{penguin3} in ST2, \emph{Volleyball} in NR and \emph{blackswan} in Davis.}
\label{img::demo}
\end{figure*}
\textbf{Parameter Settings.}
We partition each candidate region into about $600$ non-overlapping superpixels. Similar to experimental setting in work~\cite{Yeo2017Superpixel}, we extract 3-dimensional mean colors in LAB space features for each superpixel. The parameters in Eq.(\ref{eq::mode1_our}) are set empirically as $\left\{\lambda_{1} = 0.01, \lambda_{2} = 0.07 \right\}$. The parameters $\{\alpha,\beta\}$ in Eq.(\ref{eq::Fusion Model})) are set to $\left\{0.001, 50 \right\}$. In Eq.(\ref{eq::eq1}), we set the scaling parameter $ \sigma = 10 $. \\
\textbf{Evaluation Metrics.} Similar to many other tracking works, our tracking results are measured based on two evaluation metrics, i.e., distance precision rate and overlap success rate~\cite{Henriques2015High,OTB2015}.
For precision rate, it is defined to measure the difference between the estimated position and the center position of the ground-truth. The frame whose center distance is less than a certain threshold is regarded as the correctly tracked frame and participates in the precision calculation. The final precision is determined by the ratio of correct frames w.r.t total frames. We set the threshold to 20 pixels.
For overlap success rate, the average intersection over union (IoU) is used to measure overlap between the tracking result and the ground-truth. We calculate the segmentation overlap ratio (average IoU ratios of segmentation masks and ground-truth masks) and bounding box overlap ratio (average IoU ratios of ground-truth and estimated bounding boxes), respectively.
The results of comparison methods~\cite{Yeo2017Superpixel,OGBDT2015,Godec2011Hough,Shu2011Superpixel,Duffner2014PixelTrack,Hong2015MUlti,Danelljan2014Accurate,MEEM2014} have been reported in previous works~\cite{Yeo2017Superpixel,Tracking-by-Segmentation2018} and we use them in Table~\ref{table::table2}.
\subsection{Performance Evaluation}
We compare the proposed model against three traditional tracking-by-detection methods, namely MUSTer~\cite{Hong2015MUlti}, DSST~\cite{Danelljan2014Accurate} and MEEM~\cite{MEEM2014}, and five segmentation-based tracking algorithms including AMCT~\cite{Yeo2017Superpixel}, OGBDT~\cite{OGBDT2015}, HT~\cite{Godec2011Hough}, SPT~\cite{Shu2011Superpixel}, PT~\cite{Duffner2014PixelTrack}.\\
Table~\ref{table::table1} reports the results of average IoU ratios of segmentation masks and ground truth masks on all datasets. We can note that the proposed method exceeds some other tracking-by-segmentation methods, especially on ST2 data-set~\cite{Li2013Video}. This clearly demonstrates the effectiveness of the proposed GCN tracking model by further incorporating structural information into superpixel feature representation and thus can obtain more accurate segmentation results.
The proposed method can also obtain better performance than other methods on some larger datasets, such as DAVIS~\cite{Perazzi2016A}.
Table~\ref{table::table2} shows average IoU ratios of ground truth and estimated bounding boxes on different datasets. The first column in the table illustrates that our method not only outperforms existing tracking-by-segmentation methods, but also obviously performs better than some other tracking-by-detection methods. The comparsion results of these methods have been reported in ~\cite{Yeo2017Superpixel}, here, we use them.
We also compare our method with AMCT~\cite{Yeo2017Superpixel} on the area under curve (AUC) which is most related with our approach.
Figure~\ref{img::plots} (a) and (b) show the success rate in terms of bounding box and segmentation mask, respectively.
Figure~\ref{img::plots} (c) shows the precision plots in terms of bounding box.
Figure~\ref{img::demo}
It can be intuitively obtained from Figure~\ref{img::plots}, bounding box overlap ratio performance exceeds AMCT~\cite{Yeo2017Superpixel} by 1.8\% and 2.6\% higher in the segmentation overlap ratio. Compared with AMCT~\cite{{Yeo2017Superpixel}}, the proposed method can gain the precision of the bounding boxes gains of 2.7\%. Figure~\ref{img::demo} shows some visualization results.
\begin{table}[h]
\small
\centering
\caption{Comparison of the proposed algorithm with only-smoothing, and none of both on overall datasets.}
\label{table::table3}
\vspace{1 ex}
\setlength{\tabcolsep}{1.0mm}{
\begin{tabular}{l|c|c|c}
\hline
\hline
& Ours & only-smoothing & none \\
\hline
{Success-Seg} & {\bf 66.6} & 65.6 & 64.6 \\
{Success-Box} & {\bf 69.3} & 68.2 & 67.1 \\
{Precision} & {\bf 68.3} & 68.1 & 66.7 \\
\hline
\hline
\end{tabular}}
\end{table}
\subsection{Ablation study}
In this section, we will focus on the detailed analysis of the proposed GCN feature learning module. The general graph convolution is the same as Laplacian smoothing.
As shown in table~\ref{table::table3}, we explore the effects of Laplacian smoothing and our moudle on the experiment respectively. Among them, only-smoothing means that the features are only learned by employing one layer graph Laplacian smoothing convolution. 'none' means that we do not perform any graph convolutional operations on the original input features. 'Success-Seg', 'Success-Box' and 'Precision' correspond to the three evaluation results in Figure~\ref{img::plots}.
Here, we can note that, the proposed graph mixed convolutional network model performs more effectively than purely using smoothing based graph convolution. That is because the proposed GCN model fuses the two convolutional operations and thus fully considers the discriminability and similarity of the features. The best results are shown in bold in the table~\ref{table::table3}, which obviously validates the effectiveness of the proposed model.
\section{Conclusion}
This paper proposes a simple and efficient graph convolutional representation model for object segmentation-based tracking tasks. The proposed model provides a general end-to-end framework which integrates label linear prediction and structure-aware feature information of each super-pixel together to obtain object segmentation and further improve the performance of tracking. An effective algorithm has been developed to optimize the proposed model. Extensive experiments show that our method obtains better performance.
{\bibliographystyle{elsarticle-num}}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 8,113
|
Q: One-liner Perl command to rename files So far this one-liner is stripping off one line and renaming the file, but I need help to alter it so that it strips that line I am looking for Data for and remove the old file extension .csv instead of adding to it. (.csv.out). I am not sure if this can be done with one-liner.
Instead it's adding on the the extension filename.csv.out
Example
test_20110824.csv.out
One-liner:
find -type f -name '*.csv' -exec perl -i.out -wlne '/^Data for/ or print' {} \;
I want to replace the extension:
test_20110824.out
A: perl -MFile::Copy -we 'for (glob "*.csv") { my ($name) = /^(.+).csv/i; move($_, $name . ".out"); }'
To remove the header matching Data for:
perl -MFile::Copy -MTie::File -wE 'for (glob '*x.csv') { tie my @file,
"Tie::File", $_ or die $!; shift @file if $file[0] =~ /^Data for/;
untie @file; my ($name) = /^(.*).csv/i; move($_, $name . ".out"); }'
But then it's really not a one-liner anymore...
use strict;
use warnings;
use Tie::File;
use File::Copy;
use autodie;
for (@ARGV) {
tie my @file, "Tie::File", $_;
shift @file if $file[0] =~ /^Data for/;
untie @file;
my ($name) = /^(.*).csv/i;
move($_, $name . ".out");
}
And use with:
$ script.pl *.csv
A: A simple Bash shell script will suffice
(shopt -s failglob; for i in *.csv.out; do echo mv $i ${i%csv.out}out; done)
The shopt -s failglob is needed to ensure that if there are no matches the command will fail instead of trying to rename *.csv.out to *.out. The construct ${i%csv.out}out removes a trailing csv.out and replaces it with just out.
As I have coded it here, this will just echo the commands it would execute. When you're satisfied it does what you want, remove the word echo.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 4,099
|
THE 1999 PARTY: Live At The Chicago Auditorium 1974 1997
01. Intro/ Standing On The Edge (4:17)
02. Brainbox Pollution (7:52)
03. It's So Easy (11:03)
04. You Know You're Only Dreaming (4:43)
05. Veterans Of 1000 Psychic Wars (2:21)
06. Brainstorm (9:19)
07. Seven By Seven (9:26)
01. The Watcher (6:40)
02. The Awakening (2:40)
03. Paradox (5:43)
04. You'd Better Believe It (8:09)
05. The Psychedelic Warlords (3:47)
06. D-Rider (7:46)
07. Sonic Attack (4:31)
08. Master Of The Universe (6:57)
09. Welcome To The Future (2:32)
Dave Brock: Vocals, Guitar
Nik Turner: Saxophone, Flute, Vocals
Simon King: Drums
Del Dettmar: Synthesisers
Lemmy: Bass, Vocals
Simon House: Keyboards
Stacia: Mime & Dance
Recorded live at the Chicago, Auditorium Theater, USA 21/03/74
(P) © 1997 EMI Records Ltd.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,089
|
{"url":"http:\/\/crypto.stackexchange.com\/tags\/block-cipher\/hot?filter=month","text":"# Tag Info\n\n## Hot answers tagged block-cipher\n\n9\n\nThe fastest block cipher is identity, which leaves input blocks completely unchanged. This is infinitely fast on all platforms; however, it is not secure. So maybe you want the fastest block cipher that still offers some given non-trivial level of security? Then it depends a lot on what you want to implement the block cipher on. With recent PC, you would ...\n\n7\n\nIs Rijndael the fastest block cipher in the world? No. On an Intel 64 Sandy Bridge without AES-NI, AES (a subset of Rijndael) is outperfomed by ChaCha20 (and also likely by Threefish 512 which has about 6-7cpb cost on an older Intel Core 2 Duo with 64-bit ASM (link: original Skein paper PDF)) as opposed to AES' 11 cpb. (7.59 cpb on an Intel Core 2) ...\n\n3\n\nIn general you want to treat primitives like block ciphers as black boxes. You first analyze and try to break the block cipher. Once it is proven to operate correctly you can use it as primitive for a block cipher mode of operation. The mode of operation can then be proven to be secure assuming that the block cipher primitive operates well. If you don't ...\n\n2\n\nGOST symmetric cipher uses 256-bit keys to encrypt 64-bit blocks. That means there are many keys which give the same result This is actually incorrect. A block cipher, when given a key, maps $n$-bit plaintext blocks to $n$-bit ciphertext blocks (and vice versa for decryption). Here's a toy example for $n=2$: $E_k(00) = 01 \\\\ E_k(01) = 10 \\\\ E_k(10) ... 2 This is because the set of possible permutations of 64 bit blocks of plaintext ($2^{64}$possibilities) to 64 blocks of ciphertext is very high. A key selections just one of these permutations. Even a 256 bit key space is smaller by far than the number of possible permutations. Some plaintext blocks will likely map to the same ciphertext block for a few of ... 2 If you are going to generate ALL the numbers in the output domain, an attacker's job of guessing the next number becomes easier the more numbers you use, once you pass the halfway point. I would suggest with whatever method you choose, to never exceed$N\/2$values used, where$N\\$ is the total number of elements that can be generated using that method, ...\n\n1\n\nIn symmetric cryptography it is hard to prove security properties on algorithm. Most of block ciphers relies on showing resistances to the current attacks (cf the paper you linked or any paper that introduce a new block cipher). As nobody can know what will be the next attack vector, it is not possible to be prepared against it. From The design of Rijndael ...\n\n1\n\nThe speed of a cipher actually depends on lots of factors, including: The specific hardware platform you're considering (CPU architecture, instruction set, number of cores etc). Implementation details. Compiler flags used. Some ciphers have a large initial overhead due e.g. to a slow key setup; as a result they are slow when encoding very small messages. ...\n\n1\n\nAssuming that there are no collisions and there exists a 1 to 1 bijection between input and output, you could construct and store a table of assignments on a single round of encryption for all 256 possible bytes regardless of the key size. From there you can extend this to subsequent rounds by simply XORing the result of the previous round to the next 8 bit ...\n\n1\n\nCheck out Swap or Not (pdf). Unlike some Feistel-network based solutions, this will provide you with near-ideal security (the adversary would have to query close to the entire space to have non-negligible advantage). Alternatively, enumerating and shuffling a list of 2^16 16-bit numbers would require only ~128KB of RAM. If you needed to reproduce the ...\n\n1\n\nDo I really need a random IV if I salt the data? Prepending 128-bits of random salt to the data is the same as using a 128-bit random IV, right? Yes, it should do the job, but if I'm reading your question correctly you don't really have to optimize storage away, but if you want to, I'd rather recommend going with AES-128 and a 128-bit IV which is ...\n\n1\n\nHere is how I would approach this: First off, strip off the unkeyed parts of the cipher at the beginning and the end. That is, process the plaintexts with the 'rotr 8\/add\/rotl 3' at the beginning, and process the ciphertexts with the 'xor\/rotr 3' at the end (rotr because we're working the inverse direction). Next, we focus in on only the right side ...\n\nOnly top voted, non community-wiki answers of a minimum length are eligible","date":"2016-05-04 00:01:37","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.2611916959285736, \"perplexity\": 1288.0155434998576}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.3, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2016-18\/segments\/1461860121985.80\/warc\/CC-MAIN-20160428161521-00086-ip-10-239-7-51.ec2.internal.warc.gz\"}"}
| null | null |
{"url":"https:\/\/zbmath.org\/?q=an:1209.62060","text":"# zbMATH \u2014 the first resource for mathematics\n\n##### Examples\n Geometry Search for the term Geometry in any field. Queries are case-independent. Funct* Wildcard queries are specified by * (e.g. functions, functorial, etc.). Otherwise the search is exact. \"Topological group\" Phrases (multi-words) should be set in \"straight quotation marks\". au: Bourbaki & ti: Algebra Search for author and title. The and-operator & is default and can be omitted. Chebyshev | Tschebyscheff The or-operator | allows to search for Chebyshev or Tschebyscheff. \"Quasi* map*\" py: 1989 The resulting documents have publication year 1989. so: Eur* J* Mat* Soc* cc: 14 Search for publications in a particular source with a Mathematics Subject Classification code (cc) in 14. \"Partial diff* eq*\" ! elliptic The not-operator ! eliminates all results containing the word elliptic. dt: b & au: Hilbert The document type is set to books; alternatively: j for journal articles, a for book articles. py: 2000-2015 cc: (94A | 11T) Number ranges are accepted. Terms can be grouped within (parentheses). la: chinese Find documents in a given language. ISO 639-1 language codes can also be used.\n\n##### Operators\n a & b logic and a | b logic or !ab logic not abc* right wildcard \"ab c\" phrase (ab c) parentheses\n##### Fields\n any anywhere an internal document identifier au author, editor ai internal author identifier ti title la language so source ab review, abstract py publication year rv reviewer cc MSC code ut uncontrolled term dt document type (j: journal article; b: book; a: book article)\nGlobal uniform risk bounds for wavelet deconvolution estimators. (English) Zbl\u00a01209.62060\nSummary: We consider the statistical deconvolution problem where one observes $n$ replications from the model $Y=X+\\varepsilon$, where $X$ is the unobserved random signal of interest and $\\varepsilon$ is an independent random error with distribution $\\varphi$. Under weak assumptions on the decay of the Fourier transform of $\\varphi$, we derive upper bounds for the finite-sample sup-norm risk of wavelet deconvolution density estimators $f_n$ for the density $f$ of $X$, where $f: \\Bbb R \\rightarrow \\Bbb R$ is assumed to be bounded. We then derive lower bounds for the minimax sup-norm risk over Besov balls in this estimation problem and show that wavelet deconvolution density estimators attain these bounds. We further show that linear estimators adapt to the unknown smoothness of $f$ if the Fourier transform of $\\varphi$ decays exponentially and that a corresponding result holds true for the hard thresholding wavelet estimator if $\\varphi$ decays polynomially. We also analyze the case where $f$ is a \u201csupersmooth\u201d\/analytic density. We finally show how our results and recent techniques from Rademacher processes can be applied to construct global confidence bands for the density $f$.\n\n##### MSC:\n 62G07 Density estimation 42C40 Wavelets and other special systems 62G15 Nonparametric tolerance and confidence regions\nFull Text:","date":"2016-04-30 09:05:43","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.8091437220573425, \"perplexity\": 3636.0555600762136}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 5, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2016-18\/segments\/1461860111809.10\/warc\/CC-MAIN-20160428161511-00049-ip-10-239-7-51.ec2.internal.warc.gz\"}"}
| null | null |
THE MOST DISTINGUISHED MINERVA PRESS NOVELIST
[BAGE, Robert].
Hermsprong; or, Man as he is not. A Novel … By the Author of Man as he is.
London: Printed for William Lane, at the Minerva Press … 1796.
Three vols., 12mo., lacking half-titles but with four pages of advertisements at rear of vol. I (including a long review of Man as he is), and single leaf of advertisements at rear of vol. II; small hole in L6, vol. III, loss of one letter; contemporary tree calf, morocco lettering and numbering-pieces; slight crease to back cover of vol. III, slight cracks to joints but not weak, a very good copy.
First edition of Bage's last and finest novel.
In Hermsprong, Bage contrasts the deficiencies of English society with the beauties of the utopian community among the 'aborigines' of North America. 'There is occasionally a little tincture of the new philosophy, as it is called, and a shade of gloom is thrown upon human life' (Critical Review); but his philosophical tendencies never obscure his powerful characterisation and style. The plot turns on the wooing of a peer's only daughter by an American 'incognito' who settles in Cornwall.
Despite a considerable contemporary reputation, Bage remains one of the lesser-known novelists of the revolutionary and philosophical school led by William Godwin and Thomas Holcroft. Strongly influenced by Voltaire, Paine, and Rousseau, Bage used his novels to express his ideas about social equality and the rights of man. But Hermsprong is also 'rich in that half-acid, half-tolerant revelation of the permanent foibles of human nature in which Bage anticipated Jane Austen ... [Bage's] sound judgement of character, and the pleasant irony of his style, give him at least a place in the company of Fielding, Austen, and Thackeray' (Blakey, p. 65).
Black, The Epistolary Novel, 357; Garside, Raven and Schöwerling 1796:21.
UTOPIA MORE, Thomas.
The Common-wealth of Utopia: containing a learned and pleasant Discourse of the best State of a publike-Weale, as it is found in the Government of the new Ile called Utopia …
Fifth edition of More's Utopia in English, translated by Ralph Robinson – the last edition of his translation, first published in 1551, and revised in 1556. Alsop printed a corrected edition in 1624, with a dedication to More's grandson, Cresacre More, which is reprinted here.
[JOHNSON, Samuel].
The Prince of Abissinia. A Tale. In two Volumes …
First edition of Johnson's only novel, written in the evenings of a single week to pay for his mother's funeral. Its rapid execution is said to have been due to the fact that he had been pondering its chief topics all his life. It soon became his most popular work. Although now inevitably called 'Rasselas' after the name of the hero, that title was not used in the author's lifetime except for the first American edition (1768).
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,888
|
\section{Introduction}
Ancestral sequence reconstruction (ASR) is a key task in computational evolutionary biology~\cite{liberles_ancestral_2007}. It consists in inferring a molecular sequence at an ancestral species of a known phylogeny, given descendant sequences at the tip of the tree. Numerous approaches are available for this task. Some are based on statistical models of sequence evolution on a tree, while others rely on combinatorial optimization formulations~\cite{semple_phylogenetics_2003, yang_molecular_2014}. In addition to its many biological applications, ASR has played a key role in elucidating the statistical performance of phylogeny estimation methods~\cite{mossel_impossibility_2003,mossel_phase_2004,roch_toward_2010,roch_phase_2017}. Here we establish a formal connection to sequence alignment.
Rigorous analyses of the accuracy of ASR methods have been performed mainly in two asymptotic settings. In phylogenies of arbitrarily large depth, an achievable goal is to infer a sequence that is correlated site-by-site with the true ancestral sequence~\cite{steel_five_1995,ioffe_extremality_1996,mossel_recursive_1998,evans_broadcasting_2000}. In the taxon-rich setting, on the other hand, where the depth of the phylogeny is bounded as the number of taxa increases, consistent estimators are known to exist~\cite{gascuel_inferring_2010,roch_sufficient_2021}. That is, under conditions on the branching of the phylogeny around its root, the correct inference of a single site in the ancestral sequence can be guaranteed as the number of leaves goes to infinity.
Most theoretical results in this area are derived under models of sequence evolution by single site substitutions. More complex models allowing for site insertions and deletions (indels) have also been considered~\cite{andoni_global_2012,ganesh_optimal_2019,fan_statistically_2020}. The star case, also known as trace reconstruction, has been the subject of much recent interest~\cite{holenstein_trace_2008,nazarov_trace_2017,holden_lower_2020,davies_approximate_2021,davies_reconstructing_2019}. See also \cite{THATTE200658,mitrophanov_convergence_2007,daskalakis2013alignment,Allman2015StatisticallyCK,fan2020impossibility} for rigorous analyses of indel models in other contexts, e.g., distance-based phylogeny reconstruction.
Indel models are closely related to another important bioinformatics problem, multiple sequence alignment (MSA), in which one attempts to best align a collection of molecular sequences under some mismatch penalty score by inserting gaps. In practice, MSA is a hard problem, especially at large evolutionary distances~\cite{rost_twilight_1999,chang_phylogenetic_2008}. While statistical approaches based on indel models have also been developed~\cite{Lunter2005}, commonly used approaches involve progressively aligning the given sequences up a guide tree, in what is reminiscent of ASR procedures~\cite{ranwez:hal-02535389}. In fact, many trace reconstruction and ASR methods under indels involve partial local alignments of sequences.
In this paper, we combine insights from ASR in the taxon-rich setting together with the probabilistic analysis of indel models to prove the first (as far as we know) rigorous guarantee for sequence alignment under an indel model on a phylogenetic tree. Our result is somewhat counter-intuitive: we show that perfect pairwise sequence alignment with high probability is in principle possible \emph{at arbitrary large evolutionary distances}---provided the phylogeny is known and dense enough. While such a condition may not be satisfied in real datasets, our analysis is a step towards a better theoretical understanding of MSA and its connections to ASR.
In a nutshell, we take advantage of the density of the phylogeny to estimate ancestral sequences with high probability along the path between two leaf sequences of interest, then reconstruct the history of mutations along the way. For the ASR step, we use a standard phylogenetic method known as parsimony, which seeks to use the smallest number of mutations possible to explain sequences at the leaves of a phylogeny. Rigorous analyses of parsimony are often challenging and have revealed the intricate, often unintuitive, behavior of the method~\cite{li_more_2008,fischer_maximum_2009,zhang_analyzing_2010,herbst_ancestral_2017,herbst_accuracy_2018}. In our taxon-rich setting, branching process results lead to rigorous guarantees on the ancestral reconstruction.
The rest of the paper is organized as follows. In Section~\ref{section:main-results}, we state our main result after introducing some background. The alignment algorithm is presented in Section~\ref{section:alignment}. The proof is comprised of two parts: the ancestral estimation step is analyzed in Section~\ref{section:ancestral} and the alignment step is analyzed in Section~\ref{section:one-mutation}.
\section{Background and main result}
\label{section:main-results}
In this section, we state our main result. First, we introduce the model of sequence evolution we use here as well as the multiple sequence alignment problem.
\subsection{Definitions}
We consider the TKF91 insertion-deletion (indel) sequence evolution model. Technically, we use a slight variant of the TKF91 model defined in~\cite{Thorne1991}, where we only allow an alphabet with two letters $0$ and $1$ to simplify the analysis and its presentation. Our results extend naturally to more general settings.
\begin{definition}[TKF91 model: two-state version]
\label{Def:BinaryIndel}
Consider the following Markov process $\mathcal{I} = \{\mathcal{I}_{t}\}_{t \geq 0}$ on the space $\mathcal{S}$ of binary digit sequences together with an \textbf{immortal link $``\bullet"$}, that is,
\begin{equation*}\label{S}
\mathcal{S} := ``\bullet" \otimes \bigcup_{M\geq 1} \{0,1\}^M,
\end{equation*}
where the notation above indicates that all sequences begin with the immortal link. Positions of a sequence, except for that of the immortal link, are called \textbf{sites} or \textbf{mortal links}. Let $(\eta,\lambda,\mu) \in (0,\infty)^{3}$ and $(\pi_0,\pi_1) \in [0,1]^2$ with $\pi_0 + \pi_1 = 1$ be given parameters. The continuous-time dynamics are as follows: If the current state is the sequence $\vec{x} \in \mathcal{S}$, then the following events occur independently:
\begin{itemize}
\item \emph{Substitution:} Each site
is substituted independently at rate $\eta > 0$. When a substitution occurs, the corresponding digit is replaced by $0$ and $1$ with probabilities $\pi_0$ and $\pi_1$, respectively.
\item \emph{Deletion:} Each site
is removed independently at rate $\mu$.
\item \emph{Insertion:} Each site, as well as the immortal link, gives birth to a new digit independently at rate $\lambda$. When a birth occurs, the new site is added immediately to the right of its parent site. The newborn site has digit $0$ and $1$ with probabilities $\pi_0$ and $\pi_1$, respectively.
\end{itemize}
\end{definition}
We run this process on a rooted metric tree as follows. Consider a \textit{rooted binary tree} $T = (V,E,\rho,\mathbf{t})$ with vertices $V$, edges $E$, root $\rho$, and edge lengths $\mathbf{t} = \{t_e\}_{e \in E}$ (in time units). We restrict ourselves to ultrametric trees, that is, the sum of edge lengths from root to leaf is the same for every leaf. We refer to this common quantity as the \textbf{depth} of the tree and denote it by $h$. The rooted metric tree $T$ is then indexed by all points along the edges of $T$. The root vertex has an initial sequence $\sigma_{\rho} \in \mathcal{S}$. With an initial sequence $\sigma_u \in \mathcal{S}$, the TKF91 process is recursively performed on each descending edge $e = (u,v)$ over the time interval $[0,t_e]$ to obtain another sequence $\sigma_v \in \mathcal{S}$. Processes running along descending edges of $u$ are independent, conditioned on state $\sigma_u$ at $u$. We refer to the full process as the \textbf{(two-state) TKF91 process on tree $T$}.
For any sequence $\sigma \in \mathcal{S}$, let $|\sigma|$ be the length of the sequence, and let $|\sigma|_{0}$ and $|\sigma|_{1}$ be the number of $0$'s and $1$'s in the sequence, respectively. The stationary distribution of the sequence length $|\sigma| = M$ is known~\cite{Thorne1991} to be
\begin{equation}
\label{eq:LengthStationary}
\gamma_{M} = \left(1 - \frac{\lambda}{\mu}\right)\left(\frac{\lambda}{\mu}\right)^{M}, \qquad M \in \mathbb{Z}_+,
\end{equation}
provided $\mu > \lambda$.
We assume that the root sequence $\sigma_\rho$ follows its stationary distribution. That is, $|\sigma_{\rho}|$
is distributed according to $\gamma_{M}$ and its sites are i.i.d.~in $\{0,1\}$ with respective probabilities $\pi_0$ and $\pi_1$.
Stationarity of $\sigma_{\rho}$ implies stationarity of the TKF91 process throughout the tree. We assume from now on that $\mu > \lambda$ and that stationarity holds.
\paragraph{Some notation}
Later on, we will need the following notation.
For a sequence $\sigma \in \mathcal{S}$,
let $\mathcal{S}_{s}(\sigma)$, $\mathcal{S}_{d}(\sigma)$, and $\mathcal{S}_{i}(\sigma)$ be
the sequences that differ from $\sigma$
respectively by a single substitution, a single
deletion, and a single insertion.
Observe that these sets are disjoint as
the sequence lengths in each necessarily differ.
Further, let $\mathcal{S}_1(\sigma) = \mathcal{S}_{s}(\sigma) \cup \mathcal{S}_{d}(\sigma) \cup \mathcal{S}_{i}(\sigma)$ be the sequences obtained by performing a single mutation on $\sigma$,
and define
\begin{equation}\label{eq:lambdastar}
\lambda^{\ast}(\sigma) = \sum_{\tau \in \mathcal{S}_1(\sigma)}Q(\sigma,\tau) = \lambda (|\sigma| + 1) + \mu |\sigma| + \eta \pi_1 |\sigma|_0 + \eta \pi_0 |\sigma|_1
\end{equation}
as the total rate under the TKF91 process of moving
away from $\sigma$, where $Q(\sigma,\tau)$ is the rate
at which the TKF91 process on an edge jumps from $\sigma$ to $\tau$. Formula~\eqref{eq:lambdastar} is derived formally in the appendix.
\subsection{Multiple sequence alignment}
To compare sequences descending from a common ancestor through substitutions, insertions and deletions, it is natural to attempt to align them as best as possible, that is, to construct a multiple sequence alignment.
\begin{definition}[Multiple sequence alignment]
For any integer $n \geq 1$ and sequences $\boldsymbol{\sigma} = (\sigma_{v_1},\ldots,\sigma_{v_m}) \in \mathcal{S}^m$ at points $v_1,\ldots,v_m \in T$, a \textbf{multiple sequence alignment} (or pairwise alignment when $m=2$) is a collection of sequences $\mathbf{a}(\boldsymbol{\sigma}) = (a_1(\boldsymbol{\sigma}),\ldots,a_m(\boldsymbol{\sigma}))$ whose entries come from $\{0,1,-\}$ ($-$ is called a \textit{gap}) such that: \begin{itemize}
\item the lengths satisfy $$|a_1(\boldsymbol{\sigma})| = |a_2(\boldsymbol{\sigma})| = \cdots = |a_m(\boldsymbol{\sigma})| \geq \max\{|\sigma_{v_1}|,|\sigma_{v_2}|,\ldots,|\sigma_{v_m}|\},$$
\item no corresponding entries of $a_1(\boldsymbol{\sigma}),\ldots,a_m(\boldsymbol{\sigma})$ all equal $-$, and
\item removing $-$ from $a_i(\boldsymbol{\sigma})$ yields $\sigma_{v_i}$ for all $i \in \{1,2,\ldots,m\}$.
\end{itemize} A multiple sequence alignment can be expressed as an $m \times |a_1(\boldsymbol{\sigma})|$ matrix where the rows are the sequence alignments and where no column consists of all gaps. If $m=2$, the alignment is referred to as pairwise.
\end{definition}
\noindent More generally, a multiple sequence alignment procedure may take as input further auxiliary information (beyond the sequences to be aligned), such as a tree or sequences at other points of the tree. Our alignment algorithm (see Section~\ref{section:alignment}) will indeed use additional information.
Two sites, one from one sequence and the other from another sequence, are said to be \textbf{homologous} provided they descend from a common site in their most recent common ancestral sequence \textit{only through substitutions} under the evolutionary process on the tree.
A \textbf{true} multiple sequence alignment
is one that places homologous sites in the same column and non-homologous sites in different columns. We note however that certain homology relationships are unknowable a priori: for example, if in the course of evolution a $0$ is inserted in a sequence next to another $0$, which of them descends from the ancestral $0$ is arbitrary. Here we take the convention that a repeated site is always inserted at the beginning of a run; and that similarly a repeated site is always deleted at the beginning of a run.
\subsection{Statement of main result}
The following theorem states that it is possible to construct with high probability a true pairwise alignment of the sequences at two arbitrary leaves $v$ and $w$ of a phylogeny as long as the maximal branch length is sufficiently small.
\begin{theorem}[Main Result]
\label{thm:main}
Fix $\eta,\mu,\lambda \in (0,\infty)$, the substitution, deletion, and insertion rates under the TKF91 model. There is a polynomial-time alignment procedure $A$ such that
for any tree depth $h > 0$ and any failure probability $\varepsilon > 0$, there exists a maximum branch length $t_{\textnormal{max}} := t_{\textnormal{max}}(h,\varepsilon) > 0$ such that
the following property holds. For any \textit{rooted binary tree} $T = (V,E,\rho,\mathbf{t})$ with vertices $V$, edges $E$, root $\rho$, and edge lengths $\mathbf{t} = \{t_e\}_{e \in E}$, assume that the leaves $\partial T = \{\ell_i\}_{i=1}^{n}$
are ordered from left to right in a planar realization of $T$, and let
$v = \ell_1$ and $w=\ell_n$. Then the alignment procedure applied to the sequences $\sigma_{\ell_1},\sigma_{\ell_2},\ldots,\sigma_{\ell_n}$ outputs a true pairwise alignment of $\sigma_v$ and $\sigma_w$ with probability at least $1- \varepsilon$, provided that $t_e \leq t_{\textnormal{max}}$ for all edges $e \in E$.
\end{theorem}
\noindent Note that the tree depth $h$ is arbitrary. The alignment procedure, which is described in Section~\ref{section:alignment}, takes as input leaf sequences at the leaves of $T$ as well as $T$ itself.
\paragraph{Extensions}
While we assume above that the rate of substitution is the same throughout the tree, our proof still goes through if the parameter $\eta$ is merely an upper bound on that rate across edges. Similarly, our two-state assumption and the details of the substitution model
do not play a critical role in the proof. We make these assumptions to simplify the presentation.
\section{Alignment algorithm}
\label{section:alignment}
In this section, we describe the alignment procedure of
Theorem~\ref{thm:main}. We emphasize that this algorithm is not meant to be practical, but rather serve as a proof of our main result.
\subsection{Overview of full alignment algorithm}
We introduce the following alignment algorithm $A$ which takes as input a rooted metric tree $T$, two distinguished leaves $v$ and $w$, all leaf sequences, and a pre-processing parameter $\delta_1$. We take $\delta_1$ to satisfy $t_{\textnormal{max}} \leq \delta_1 \leq h$.
The algorithm outputs a
pairwise alignment for the sequences at $v$ and $w$.
There is a unique path between $v$ and $w$ that we henceforth call the \textit{backbone}. We let $B$ be the number of \textit{non-root} vertices on the backbone. Then $v = x_1$ and $w = x_B$ and the other \textit{non-root} backbone vertices are in order $x_2,...,x_{B-1}$. For some parts of the algorithm and analysis, it will be convenient to use an alternative numbering of the backbone vertices---away from the root, numbering the left side, then numbering the right side. Specifically, let $x_1,\ldots,x_{B^-}$ be the backbone vertices on the same side of the root as $x_1$. Let $x_{B^-+1},\ldots,x_B$ be the backbone vertices on the same side of the root as $x_B$ and let $B^+$ be their number. Then we set
$$
\tilde{x}^-_i := x_{B^- - (i-1)}, \qquad i =1,\ldots, B^-
$$
and
$$
\tilde{x}^+_i := x_{B^- + i}, \qquad i = 1,\ldots, B^+.
$$
Notice in particular that $\tilde{x}^-_1$ and $\tilde{x}^+_1$ are the children of the root and $\tilde{x}^-_{B^-} = x_1$ and $\tilde{x}^+_{B^+} = x_B$.
\begin{figure}[t]
\centering
\includegraphics[scale=1.2]{Figures/AlignmentProc1-12.png}
\caption{(a) Tree $T$ with leaf sequences $\sigma_1 = \sigma_{x_1},\sigma_2= \sigma_{\ell_{2,1}},..., \sigma_{\ell_{i_2}},\sigma_B = \sigma_{x_B}$. (b) Tree $T$ with leaf sequences $\sigma_1 = \sigma_{x_1},\sigma_2 = \sigma_{x_{2,1}},..., \sigma_{\ell_{i_2}},\sigma_{3} = \sigma_{x_{3,1}},...,\sigma_{\ell_{i_3}},\sigma_B$.}
\label{fig:AlignmentProc1-12}
\end{figure}
We now describe the main steps of the algorithm. Some details will be given in the following subsections.
Figure~\ref{fig:AlignmentProc1-12} illustrates part of this algorithm at a high level.
We start with a pre-processing step.
\begin{itemize}
\item \textbf{Pre-processing: backbone sparsification.} We first construct a subtree $T'$ by pruning some backbone vertices and their descendants. Initialize $T' := T$. Then, for $o = -,+$ and for $k = 1,...,B^o-1$,
\begin{enumerate}
\item Check whether the vertex $\tilde{x}^o_k$ is a vertex in the tree $T'$. If not, do nothing.
\item\label{item:preproc-ell} If $\tilde{x}^o_k$ is in the tree $T'$, find the minimal $\ell \geq 1$ such that the distance (accounting for edge lengths) between $\tilde{x}^o_k$ and $\tilde{x}^o_{k+\ell}$ is at least $\delta_1$. Observe that, by assumption, the distance between $\tilde{x}^o_k$ and $\tilde{x}^o_{k+\ell}$ is necessarily at most $\delta_1 + t_{\textnormal{max}} \leq 2 \delta_1$.
\item Remove the vertices $\tilde{x}^o_{k+1},...,\tilde{x}^o_{k+\ell-1}$, except $\tilde{x}^o_{B^o}$, and all of their off-backbone descendants from the tree $T'$ (if they exist).
\end{enumerate}
\end{itemize}
The result is a tree where the distance between consecutive vertices on the backbone is in $[\delta_1, 2 \delta_1]$ (by the observation in Item~\ref{item:preproc-ell}), with the possible exception of the children of the root and the last pair on each side of the root all of whose distances are in $(0,2\delta_1]$. To simplify the notation, we re-assign $T$ to be this new rooted metric tree and we re-assign $x_1, x_2, \ldots,x_B$ to be the backbone vertices on this tree (with an updated value for $B$ and updated alternative numbering $\tilde{x}^o_k$ for $o = -,+$ and $k=1,\ldots,B^o$).
We then proceed with the alignment algorithm, which consists of two main steps both proceeding along the backbone:
\begin{enumerate}
\item \textbf{Ancestral estimation:} We infer the ancestral sequences at the backbone vertices as follows. For $k = 2,...,B-1$:
\begin{enumerate}
\item For the child vertex $z_k$ of $x_k$ that is off the backbone, infer the sequence $\hat{\sigma}_{z_k}$ at $z_k$ using the Fitch method~\cite{Fitch71} (described below in Section~\ref{section:fitch}) applied to the subtree rooted at $z_k$.
\item Set $\hat{\sigma}_{x_k}$ equal to $\hat{\sigma}_{z_k}$.
\end{enumerate}
\item \textbf{Recursive alignment:} Now that the sequences at the non-root backbone vertices $\{x_k\}_{k=1}^{B}$ have been estimated, we construct a multiple sequence alignment sequentially, starting from $x_1$, going to $x_2$, and ending at $x_{B-1}$ and $x_B$. This stepwise alignment procedure is described in Section~\ref{section:stepwise} below. If the inferred sequences of successive backbone vertices are not at most one mutation apart, then we terminate the algorithm with no output.
Else, a pairwise sequence alignment is produced for vertices $v = x_1$ and $w = x_B$.
\end{enumerate}
We will show in Proposition~\ref{prop:Fitch2} below that, with high probability,
$\hat{\sigma}_{z_k} = \sigma_{z_k}$ for all $k = 2,\ldots, B-1$.
We will then show in Proposition~\ref{prop:Intersection} below that the above stepwise alignment outputs a true pairwise alignment with high probability.
\subsection{Ancestral sequence reconstruction}
\label{section:fitch}
We briefly describe below the ancestral sequence reconstruction subroutine. Note that we use the Fitch method for the convenience of its analysis, but other methods could also be used.
\begin{definition}[Fitch estimator]
Let $T = (V,E)$ be a finite binary rooted tree with root $z$ and leaf set $\partial T \subset V$ with given leaf sequences $(\sigma_{\ell})_{\ell \in \partial T}$. For any leaf vertex $\ell$, define $\hat{S}_\ell \subset \mathcal{S}$ to be the subset $\hat{S}_\ell = \{\sigma_\ell\}$. For each non-leaf vertex $v$ with children $v_1$ and $v_2$, define $\hat{S}_v \subset \mathcal{S}$ recursively to be \begin{align*}
\hat{S}_v = \begin{cases}
\hat{S}_{v_1} \cap \hat{S}_{v_2} & \textnormal{if} \ \hat{S}_{v_1} \cap \hat{S}_{v_2} \ne \emptyset \\
\hat{S}_{v_1} \cup \hat{S}_{v_2} & \textnormal{otherwise}.
\end{cases}
\end{align*} Then define the \textit{Fitch estimator} $\hat{\sigma}_z$ of $\sigma_z$ to be a uniformly chosen member of $\hat{S}_z$.
\end{definition}
\noindent An analysis of this method in our setting is provided in Proposition~\ref{prop:Fitch2} below.
\subsection{Stepwise alignment}
\label{section:stepwise}
In this section, we describe the stepwise alignment subroutine. It is based on the assumption that along the backbone (of the pruned tree):
\begin{enumerate}
\item[(i)] the sequences have been correctly inferred; and
\item[(ii)] consecutive ones differ by at most one mutation.
\end{enumerate}
We establish these facts in Propositions~\ref{prop:Fitch2} and~\ref{prop:Intersection} below.
In these circumstances, we show that homologous sites can be traced
(up to the convention we described earlier).
We will construct a sequence of alignments
$\mathbf{a}^2$, $\mathbf{a}^3$, etc.
We first describe the alignment of two sequences, then the alignment of alignments, and so on.
Given two sequences $\hat\sigma, \hat\tau$ satisfying the assumptions (i) and (ii) above, there are three possible cases:
\begin{enumerate}[label=(\Alph*)]
\item If $\hat\sigma=\hat\tau$, then a true
alignment
is obtained by setting $a^{2}_1(\hat\sigma,\hat\tau) = \hat\sigma$ and $a^{2}_2(\hat\sigma, \hat\tau) = \hat\tau$, corresponding to no mutation.
\item If $|\hat\sigma| = |\hat\tau|$ but $\hat\sigma$ and $\hat\tau$ agree on all sites except one,
then a true alignment
is obtained by setting $a^{2}_1(\hat\sigma,\hat\tau) = \hat\sigma$ and $a^{2}_2(\hat\sigma,\hat\tau) = \hat\tau$, corresponding to exactly one substitution between the sequences.
\item If $|\hat\sigma| = |\hat\tau| + 1$ (or vice versa) and there exists $j \in \{1,2,...,|\hat\tau|\}$ and $\hat\sigma_{\textnormal{ins}} \in \{0,1\}$ such that \begin{align*}
\hat\sigma_i = \begin{cases}
\hat\tau_i & i < j \\
\hat\sigma_{\textnormal{ins}} & i = j \\
\hat\tau_{i-1} & i > j.
\end{cases}
\end{align*}
As we discussed before, the location of the insertion cannot be determined from the sequences alone.
For example, if $\hat\sigma$ and $\hat\tau$ are separated by an insertion so that they are given by
\begin{align*}
\hat\tau &= (0,1,0,1,0,1,0,0,0,0,0,1,0)\\
\hat\sigma &= (0,1,0,1,0,1,0,0,0,0,0,0,1,0),
\end{align*}
we cannot tell which site gave birth to the new $0$ to obtain $\hat\sigma$. So we assume by convention that $j$ is the minimal choice possible.
Then a true
alignment is obtained by setting
$a^{2}_1(\hat\sigma,\hat\tau) = \hat\sigma$ and for $i=1,\ldots,|\hat\tau|+1$
\begin{align*}
a^{2}_2(\hat\sigma,\hat\tau)_i =
\begin{cases}
\hat\tau_i & i < j \\
- & i = j \\
\hat\tau_{i-1} & i > j,
\end{cases}
\end{align*}
corresponding to
a single site $\hat\sigma_{\textnormal{ins}}$ being inserted into the sequence $\hat\tau$ to the left of the $j$th site to obtain $\hat\sigma$.
\end{enumerate}
In fact, we will need to align
alignments along the backbone, rather than sequences. Suppose we have sequences $\hat\sigma_{x_1},\hat\sigma_{x_2},...,\hat\sigma_{x_B}$ and successive pairs $\{\hat\sigma_{x_1},\hat\sigma_{x_2}\},\{\hat\sigma_{x_2},\hat\sigma_{x_3}\},...,\{\hat\sigma_{x_{B-1}},\hat\sigma_{x_B}\}$ each satisfy exactly one of the cases (A), (B), or (C). (We terminate without output if the assumptions do not hold.) Then we recursively construct a multiple sequence alignment as follows.
To simplify the notation, we let
$
\hat\sigma_{1:k}
= (\hat\sigma_{x_1},\ldots,\hat\sigma_{x_k}).
$
\begin{enumerate}
\item Given $\hat\sigma_{x_1}$ and $\hat\sigma_{x_2}$, let $a^{2}_1(\hat\sigma_{1:2})$ and $a^{2}_2(\hat\sigma_{1:2})$ be the pairwise alignment constructed above.
\item For $k = 3,...,B$: \begin{enumerate}
\item We are given a multiple alignment $a^{k-1}_1(\hat\sigma_{1:k-1}),\ldots,a^{k-1}_{k-1}(\hat\sigma_{x_{1:k-1}})$ of the sequences $\hat\sigma_{x_1},\ldots,\hat\sigma_{x_{k-1}}$, and a new sequence $\hat\sigma_{x_k}$ that is at most one mutation away from $\hat\sigma_{x_{k-1}}$.
\item The sequences $\hat\sigma_{x_{k-1}}$ and $\hat\sigma_{x_k}$ satisfy one of the three cases (A), (B) or (C) by assumption, so their alignment $a^{k}_{k-1}(\hat\sigma_{x_{1:k}})$ and $a^{k}_{k}(\hat\sigma_{1:k})$ (within the larger multiple sequence alignment) will differ by at most one entry similarly to the sequence case above. The full alignment is defined as follows:
\begin{itemize}
\item If $\hat\sigma_{x_{k-1}} = \hat\sigma_{x_k}$, then set $a^{k}_k(\hat\sigma_{1:k})$ to be equal to $a^{k-1}_{k-1}(\hat\sigma_{1:k-1})$ and $a^{k}_i(\hat\sigma_{1:k})$ to be equal to $a^{k-1}_i(\hat\sigma_{1:k-1})$ for all $i < k$.
\item If $\hat\sigma_{x_{k-1}}$ and $\hat\sigma_{x_k}$ have equal length and disagree at a single segregating site,
set $a^{k}_i(\hat\sigma_{1:k})$ to $a^{k-1}_i(\hat\sigma_{1:k})$ for all $i \leq k-1$. Each entry of $a^{k}_k(\hat\sigma_{1:k})$ is set to the corresponding entry of $a^{k}_{k-1}(\hat\sigma_{1:k})$, except for the segregating site. If the latter occurs at position $i$ within $a^{k}_{k-1}(\hat\sigma_{1:k})$, then we set $a^{k}_k(\hat\sigma_{1:k})_i$ to $a^{k}_{k-1}(\hat\sigma_{1:k})_i + 1 \ (\textnormal{mod}\ 2)$.
\item If
$\hat\sigma_{x_k}$ has one more site than $\hat\sigma_{x_{k-1}}$, then an insertion has occurred and the inserted site in $\hat\sigma_{x_k}$ cannot be ancestral to any site in $\hat\sigma_{x_1},\ldots,\hat\sigma_{x_{k-1}}$. So the inserted site in $\hat\sigma_{x_{k}}$ must correspond to a gap in all previous sequences. More specifically, if the site $\hat\sigma_{\textnormal{ins}}$ is inserted
to the left of position
$j^{\ast} \in \{1,\ldots,|a^{k-1}_{k-1}(\hat\sigma_{1:k-1})|\}$
in the $(k-1)$-st sequence \textit{in the previously constructed alignment $a^{k-1}_{k-1}(\hat\sigma_{1:k-1})$} (where $j^{\ast}$ is the minimal such choice)
then set
\begin{align*}
a^k_k(\hat\sigma_{1:k})_{i} = \begin{cases}
a^{k-1}_{k-1}(\hat\sigma_{1:k-1})_{i} & 1 \leq i < j^{\ast} \\
\hat\sigma_{\textnormal{ins}} & i = j^{\ast} \\
a^{k-1}_{k-1}(\hat\sigma_{1:k-1})_{i-1} & j^{\ast} < i \leq |a^{k-1}_{k-1}(\hat\sigma_{1:k-1})| + 1
\end{cases}
\end{align*}
and for all $\ell \leq k-1$
\begin{align*}
a^{k}_\ell(\hat\sigma_{1:k})_{i} = \begin{cases}
a^{k-1}_\ell(\hat\sigma_{1:k-1})_{i} & 1 \leq i < j^{\ast} \\
- & i = j^{\ast} \\
a^{k-1}_\ell(\hat\sigma_{1:k-1})_{i-1} & j^{\ast} < i \leq |a^{k-1}_{k-1}(\hat\sigma_{1:k-1})| + 1.
\end{cases}
\end{align*}
\item The case where
$\hat\sigma_{x_k}$ has one fewer site than $\hat\sigma_{x_{k-1}}$ is handled symmetrically.
\end{itemize}
\end{enumerate}
\item Output the pairwise alignment $(a^{B}_1(\hat\sigma_{1:B}), a^{B}_B(\hat\sigma_{1:B}))$ \emph{after removing all columns with only gaps}.
\end{enumerate}
\subsection{Theoretical guarantee}
We establish the two claims below in the next sections.
\begin{prop}[Correctness of alignment]
\label{prop:correct-align}
Let $T$ be the output of the pre-processing step and let $x_1,\ldots,x_B$ be the resulting backbone vertices. Then the alignment algorithm produces
a true pairwise alignment of $\sigma_{x_1}$ and $\sigma_{x_B}$ provided that:
\begin{enumerate}
\item (Correctness of ancestral estimation) For $k=2,\ldots,B-1$,
$\hat\sigma_{x_k} = \sigma_{x_k}$.
\item (One-mutation condition) Successive pairs of true backbone sequences
$$
\{\sigma_{x_1},\sigma_{x_2}\},\{\sigma_{x_2},\sigma_{x_3}\},\ldots,\{\sigma_{x_{B-1}},\sigma_{x_B}\},
$$
are at most one mutation away.
\end{enumerate}
\end{prop}
\section{Correctness of ancestral estimation}
\label{section:ancestral}
In this section, we analyze the ancestral sequence estimation step. The analysis proceeds by coupling the TKF91 process with a percolation process. Roughly, we say that an edge is \textbf{open} if the sequence does not change along it under the TKF91 process. We will show that, provided the edge lengths are short enough, the open cluster of the root forms a fairly ``dense'' subtree with high probability. The latter property will lead to a correct reconstruction by the Fitch method.
\subsection{The percolation process}
Consider again the backbone vertices $\{x_k\}_{k=1}^{B}$ and the off-backbone child vertices $\{z_k\}_{k=2}^{B-1}$. For each $k=2,\ldots,B-1$ separately, we couple the sequence evolution process on the subtree descending from $z_k$ with a simpler percolation process.
We will need some notation. We denote by $T_k$ the subtree of $T$ (after pre-processing) rooted at $z_k$. For two sequences $\sigma, \tau$,
we let $P_t(\sigma,\tau)$ be the probability under the TKF91 model on an edge
that, started at $\sigma$, the state is $\tau$ after time $t$. Similarly, we let $\tilde{P}_t(\sigma,\tau)$ be the same probability \textit{conditioned on not being at state $\sigma$ at time $t$}.
It will be convenient to work on an infinite tree. Specifically, let $\overline{T}_k$ be the completion of $T_k$ into an infinite binary tree where new edges have length $0$. We now describe the coupling:
\begin{itemize}
\item \textbf{The percolation process on $\overline{T}_k$.} We condition on the sequence $\sigma_{z_k}$ at $z_k$, the root of $T_k$.
For an edge $e$ on $\overline{T}_k$, let $t_e$ be its length and set
$$
\zeta^{(k)}_{e} = P_{t_e}(\sigma_{z_k}, \sigma_{z_k}),
$$
that is, the probability that the sequence does not change along edge $e$ if started at $\sigma_{z_k}$. We then perform percolation
on $\overline{T}_k$ with probabilities $\zeta^{(k)}_e$:
for each edge $e$, it is open independently with probability $\zeta^{(k)}_e$. Let $\mathcal{C}_k$ be the resulting open cluster including $z_k$ (i.e., all vertices of $\overline{T}_k$ that can be reached from $z_k$ using only open edges).
\item \textbf{The joint process on $\overline{T}_k$.} For each vertex $v$ in $\mathcal{C}_k$, set $\sigma_v = \sigma_{z_k}$. For each descendant $w$ of a vertex $v \in \mathcal{C}_k$ that is not itself in $\mathcal{C}_k$, assign a sequence to $w$ taken from the conditional distribution $\tilde{P}_{t_e}(\sigma_{z_k},\,\cdot\,)$ where $e = (v,w)$.
For each remaining vertex, we run the TKF91 process recursively from the states already assigned. Note that the edges of length $0$ added in the completion of $T_k$ simply entail copying the sequences at the leaves of $T_k$ to all their descendants.
\end{itemize}
We will be interested in the properties of the cluster $\mathcal{C}_k$.
Let $\overline{T}_k^{\mathcal{O}}$ be the subtree of $\overline{T}_{k}$ made of all the vertices in $\mathcal{C}_k$ and the edges connecting them.
For any $\ell,b \in \mathbb{Z}_{+}$, a rooted tree $T'$ is said~\cite{mossel2001} to be a \textbf{$(b,\ell)$-diluted tree} if: for all $i \in \mathbb{Z}_{+}$, each of the vertices of $T'$ at graph distance $i \ell$ from the root has at least $b$ descendants at graph distance $i\ell + \ell$ from the root.
From the following lemma adapted from \cite{mossel2001}, $\overline{T}_k^{\mathcal{O}}$ is $(2,3)$-diluted with arbitrarily high probability provided edge lengths are short enough.
\begin{figure}
\centering
\includegraphics[scale=0.75]{Figures/DilutedTree.png}
\caption{An open $2$-diluted $3$-regular subtree of the infinite binary rooted tree. Solid lines not descending from any dashed line indicate no mutation. Dashed lines indicate a mutation may have occurred. For every vertex in the even generations, at least three of its grandchildren share the same trait.}
\label{fig:DilutedTree}
\end{figure}
Figure \ref{fig:DilutedTree} depicts the infinite $2$-diluted $3$-regular tree.
Later in the proof, we will need to condition on the length of $\sigma_{z_k}$ being less than a
threshold $\bar{L}$. Let $\PP^{\bar{L}}$ be the probability measure of the joint process where $\sigma_{z_k}$ is drawn from the stationary distribution of the TKF91 process conditioned on $|\sigma_{z_k}| \leq \bar{L}$.
We first record a simple observation.
\begin{lemma}[Staying probability]
\label{lemma:staying}
Fix $\bar{L} < +\infty$.
For any sequence $\sigma$ such that $|\sigma| \leq \bar{L}$ and any $t > 0$, we have
$$
P_t(\sigma, \sigma)
\geq 1 - t (\bar{L}+1)(\mu + \lambda + \eta).
$$
\end{lemma}
\begin{proof}
Indeed
$P_t(\sigma, \sigma)$
is lower bounded by the probability
that no mutation occurs up to time $t$
which, by~\eqref{eq:lambdastar}, is at least
$$
P_t(\sigma, \sigma)
\geq \exp\left(
-(\bar{L}+1)[\mu + \lambda + \eta] t
\right)
\geq 1 - (\bar{L}+1)[\mu + \lambda + \eta] t,
$$
as claimed.
\end{proof}
\begin{lemma}[Existence of an open diluted tree]
\label{lem:Diluted}
For any $\bar{L} \in \mathbb{Z}_+$ and
$\delta_a > 0$, there is $t_{\textnormal{max}} > 0$ small enough that, if $t_e \leq t_{\textnormal{max}}$ for all $e$,
$$
\PP^{\bar{L}}[\overline{T}_k^{\mathcal{O}}\ \textnormal{is $(2,3)$-diluted}]
\geq 1 - \delta_a.
$$
\end{lemma}
\begin{proof}
By Lemma~\ref{lemma:staying}, for any sequence $\sigma_{z_k}$ such that $|\sigma_{z_k}|\leq \bar{L}$, we have that
$$
\zeta^{(k)}_{e}
= P_{t_e}(\sigma_{z_k}, \sigma_{z_k})
\geq 1
-(\bar{L}+1)[\mu + \lambda + \eta] t_{\textnormal{max}},
$$
where we recall that $t_e \leq t_{\textnormal{max}}$ by assumption (and that of course includes the added edges of length $0$).
Hence $\zeta^{(k)}_{e}$ can be made arbitrarily close to $1$ (uniformly in $e$) by taking $t_{\textnormal{max}}$ small enough
(as a function of $\bar{L}$).
The result then follows directly from \cite[Lemma 8]{mossel2001} (which can be extended in a straightforward manner to the case where percolation probabilities vary across edges but are uniformly bounded).
\end{proof}
\subsection{Analyzing the Fitch estimator}
Next, we analyze the Fitch estimator
in the event that $\overline{T}_k$ contains an open $(2,3)$-diluted subtree.
For any $D \in \mathbb{Z}_+$, let $\overline{T}_{k,D}$
be the truncation of $\overline{T}_{k}$ at level $D$, that is, the finite tree obtained by removing all vertices of $\overline{T}_{k}$ at graph distance greater than $D$ from its root. Let $\beta_k$ be the smallest positive integer such that
$\overline{\overline{T}}_k := \overline{T}_{k,2 \beta_k}$ contains all of $T_k$.
Importantly, we make the following observation about the Fitch estimator.
\begin{lemma}[Fitch estimator on the completion]
\label{lemma:fitch-completion}
The Fitch estimator applied to the leaves of $\overline{\overline{T}}_k$ produces the same ancestral sequence estimate as the Fitch estimator applied to the leaves of $T_k$.
\end{lemma}
\begin{proof}
All leaves $\bar{\bar{\ell}}$ of $\overline{\overline{T}}_k$ descending from a leaf $\ell$ of $T_k$ satisfy $\sigma_{\bar{\bar{\ell}}} = \sigma_\ell$, so by definition of the Fitch estimator $\hat{S}_\ell = \sigma_\ell$. The claim follows.
\end{proof}
Let $\overline{\overline{T}}_k^{\mathcal{O}}$ be
the truncation of $\overline{T}_k^{\mathcal{O}}$ at level $2 \beta_k$.
\begin{lemma}[Fitch estimator in the presence of an open diluted tree]
\label{lemma:fitch-diluted}
If $\overline{\overline{T}}_k^{\mathcal{O}}$
is $(2,3)$-diluted, then the Fitch estimator $\hat{\sigma}_{z_k}$ over the tree $\overline{\overline{T}}_k$
equals the true sequence $\sigma_{z_k}$
at $z_k$.
\end{lemma}
\begin{proof}
We prove this claim by induction on $\beta_k$. We start with the $\beta_k = 1$ case. Then $\overline{\overline{T}}_k$
consists of $z_k$, two children $z_k^1$ and $z_k^2$, and the grandchildren $z_k^{1,1},z_k^{1,2},z_k^{2,1},z_k^{2,2}$. If all four grandchildren belong to $\overline{\overline{T}}_k^{\mathcal{O}}$,
then we are done.
The other case, without loss of generality, is $z_k^{2,1} \notin \overline{\overline{T}}_k^{\mathcal{O}}$. Then $\sigma_{z_k^{2,1}} \ne \sigma_{z_k^{2,2}} = \sigma_{z_k}$, so the Fitch method gives $\hat{S}_{z_k^{2}} = \hat{S}_{z_k^{2,1}} \cup \hat{S}_{z_k^{2,2}} = \{\sigma_{z_k^{2,1}},\sigma_{z_k}\}$. Since $\sigma_{z_k^{1,1}} = \sigma_{z_k^{1,2}} = \sigma_{z_k}$, we have $\hat{S}_{z_k^{1}} = \{\sigma_{z_k}\}$. Continuing on, we have $\hat{S}_{z_k} = \hat{S}_{z_k^{1}} \cap \hat{S}_{z_k^2} = \{\sigma_{z_k}\}$. Since $\hat{S}_{z_k}$ contains only the state $\sigma_{z_k}$, the Fitch method is guaranteed to return $\sigma_{z_k}$.
Now, we assume the $r$-th case holds for $r \geq 1$ and we show that the $(r+1)$-st case holds as well. As before, consider the four grandchildren of $z_k$ and the same cases. If all four grandchildren belong to $\overline{\overline{T}}_k^{\mathcal{O}}$,
then they are each the root of a subtree of $2r$ levels with root state equal to $\sigma_{z_k}$. The induction assumption implies that the Fitch method returns $\sigma_{z_k}$ as estimates for $\sigma_{z_k^{i,j}}, i,j \in \{1,2\}$. The Fitch method then returns $\hat{S}_{z_k} = \{\sigma_{z_k}\}$, as required.
For the other case when $z_k^{2,1} \notin \overline{\overline{T}}_k^{\mathcal{O}}$,
we know only that $\hat{S}_{z_k^{2,1}}$
is an arbitrary set of sequences.
If $\sigma_{z_k} \in \hat{S}_{z_k^{2,1}}$,
then $\hat{S}_{z_k^{2}} = \hat{S}_{z_k^{2,1}} \cap \hat{S}_{z_k^{2,2}} = \{\sigma_{z_k}\}$,
and we are done.
Else, we have $\hat{S}_{z_k^2} = \hat{S}_{z_k^{2,1}} \cup \hat{S}_{z_k^{2,2}}$, where $\sigma_{z_k} \in \hat{S}_{z_k^{2,2}}$ so that $\hat{S}_{z_k} = \hat{S}_{z_k^1} \cap \hat{S}_{z_k^2} = \{\sigma_{z_k}\}$, as required.
This completes the proof for the $(r+1)$-st case,
and hence of the lemma.
\end{proof}
Combining Lemmas~\ref{lem:Diluted}, \ref{lemma:fitch-completion},
and
\ref{lemma:fitch-diluted},
we get the following.
\begin{prop}[Correctness of ancestral estimation off the backbone]
\label{prop:Fitch2}
For any $\bar{L} \in \mathbb{Z}_+$ and
$\delta_a > 0$, there is $t_{\textnormal{max}} > 0$ small enough that,
under $\PP^{\bar{L}}$, the Fitch estimator
on $T_k$ returns the correct ancestral state $\hat\sigma_{z_k} = \sigma_{z_k}$ with probability
at least $1-\delta_a$.
\end{prop}
\section{One-mutation condition}
\label{section:one-mutation}
In this section, we establish the one-mutation
condition required by Proposition~\ref{prop:correct-align}
and use it to finish the proof of the main result.
\subsection{A bound on the transition probabilities}
We will need a bound on the probability that
at most one mutation occurs on an edge
along the backbone. Because the state space
of the sequence process is infinite, the rates are unbounded and we state
the next bound explicitly in terms of the length of the sequence
at the start of the edge. Later on, we will use the fact that the length is stationary to control it.
\begin{lemma}[At most one mutation]
\label{lemma:atmostone}
Fix $\bar{L} < +\infty$.
For any sequence $\sigma$ such that $|\sigma| \leq \bar{L}$ and any $t > 0$, we have
$$
P_t(\sigma, Y_\sigma)
\geq 1 - \left\{t (\bar{L}+2)[\mu+\lambda+\eta]\right\}^2,
$$
where $Y_{\sigma} = \{\sigma\} \cup \mathcal{S}_1(\sigma)$ are the sequences at most one mutation away from $\sigma$.
\end{lemma}
\begin{proof}
For a TKF91 process on an edge started at $\sigma$, let $X_s \in \mathcal{S}$ be the sequence observed at time $s \in [0,t]$ and $T_i$ be the time of the $i$th jump from one state to another state. Then
$$
P_t(\sigma, Y_\sigma)
\geq \PP_\sigma[T_2 > t],
$$
as the event on the right-hand side guarantees a single
jump, which in turn guarantees that $X_t \in Y_\sigma$. Here $\PP_\sigma$ indicates that the
edge process is started at $\sigma$. Letting
$$
f_{T_1|\sigma}(s)
= \lambda^{\ast}(\sigma) \exp\left(-s \lambda^{\ast}(\sigma)\right),
\qquad
F_{T_1|\sigma}(s)
= 1 - \exp\left(-s \lambda^{\ast}(\sigma)\right),
$$
be the probability density function and cumulative distribution function of the time of the first jump
started at $\sigma$, we get by the strong Markov property
\begin{align*}
\PP_\sigma[T_2 \leq t]
&= \int_{0}^t
f_{T_1|\sigma}(s)
\sum_{\tau \in \mathcal{S}_1(\sigma)}
\frac{Q(\sigma,\tau)}{\lambda^{\ast}(\sigma)}
F_{T_1|\tau}(t-s) \,\mathrm{d} s\\
&\leq \int_{0}^t
\lambda^{\ast}(\sigma) \exp\left(-s \lambda^{\ast}(\sigma)\right)
\max_{\tau \in \mathcal{S}_1(\sigma)}
\left\{1- \exp\left(-(t-s) \lambda^{\ast}(\tau)\right)\right\}
\,\mathrm{d} s.
\end{align*}
Under the assumption that $|\sigma| \leq \bar{L}$,
it holds that $|\tau| \leq \bar{L}+1$
for any $\tau \in \mathcal{S}_1(\sigma)$,
and hence $\max\{\lambda^{\ast}(\sigma), \lambda^{\ast}(\tau)\} \leq (\bar{L}+2)[\mu+\lambda+\eta]$. Continuing on,
the last line in the previous display is
\begin{align*}
&\leq \left\{1- \exp\left(-t (\bar{L}+2)[\mu+\lambda+\eta]\right)\right\}
\int_{0}^t
\lambda^{\ast}(\sigma) \exp\left(-s \lambda^{\ast}(\sigma)\right)
\,\mathrm{d} s\\
&= \left\{1- \exp\left(-t (\bar{L}+2)[\mu+\lambda+\eta]\right)\right\}
\left\{1 - \exp\left(-t \lambda^{\ast}(\sigma)\right)\right\}\\
&\leq \left\{t (\bar{L}+2)[\mu+\lambda+\eta]\right\}^2,
\end{align*}
establishing the claim.
\end{proof}
\subsection{Union bound over the backbone}
We define a number of events whose joint occurrence
guarantees the success of our alignment procedure:
\begin{itemize}
\item \textit{(One-mutation condition)} For $o = -,+$ and $k = 1,\ldots,B^o-1$, let $F^o_k$ be the event that the sequences at $\tilde{x}^o_k$ and the backbone child vertex of $\tilde{x}^o_k$ (i.e., $\tilde{x}^o_{k+1}$) satisfy constraints (A), (B), or (C) from Section~\ref{section:stepwise}.
\item \textit{(Ancestral reconstruction)} For $o = -,+$ and $k = 1,\ldots,B^o-1$, let $G^o_k$ be the event that there is no mutation between the sequences at $\tilde{x}^o_k$ and its off-backbone child vertex $\tilde{z}^o_k$ \textit{and} that $\sigma_{\tilde{z}^o_k}$ is correctly reconstructed by applying the Fitch method on the subtree rooted at $\tilde{z}^o_k$.
\item \textit{(Root segment)} For $o = -,+$, let $H^o$ be the event that the sequences at the root and at $\tilde{x}^o_1$ are identical.
\end{itemize}
The following proposition provides a requirement on the maximum branch length $t_{\textnormal{max}}$ for all the above events to occur simultaneously. Define the bad event
$$
\mathcal{B}
= (H^-)^c \cup (H^+)^c \cup \left\{\bigcup_{o=-,+} \bigcup_{k=1}^{B^o-1} (F^o_k)^c \cup (G^o_k)^c\right\}.
$$
Recall that the pre-processing procedure has a parameter $\delta_1$.
\begin{prop}[Union bound over the backbone]
\label{prop:Intersection}
Fix a tree height $h > 0$. For any $0 < \delta_1 < h$, there is a $t_{\textnormal{max}}$ small enough
that
$$
\PP\left[ \mathcal{B} \right] \leq C h \delta_1 \log^2(\delta_1^{-1}),
$$
where $C$ is a constant depending only on $\lambda,\mu,\eta$.
\end{prop}
\begin{proof}
We take a union bound over the events making up $\mathcal{B}$.
\paragraph{Controlling the lengths} For each event, we first apply the law of total probability to control for the length of the starting sequence as follows. Suppose that sequence $\tau$ is stationary,
which we denote by $\tau \sim \Pi$. Using the stationary distribution for the length (i.e.,~\eqref{eq:LengthStationary}), we have
$$
\PP_{\tau \sim \Pi} \left[|\tau| > \bar{L}\right]
= \sum_{M=\bar{L}+1}^{\infty}
\left(1 - \frac{\lambda}{\mu}\right) \left(\frac{\lambda}{\mu}\right)^{M}
= \left(\frac{\lambda}{\mu}\right)^{\bar{L} + 1}.
$$
The expression on the right is made less than $\delta_1^2$ by choosing
\begin{equation}\label{eq:barLdef}
\bar{L}
= \bigg\lceil \frac{\log(\delta_1^{2})}{\log(\lambda/\mu)}\bigg\rceil
\leq C' \log(\delta_1^{-1}),
\end{equation}
for a constant $C' > 0$ depending only on $\mu, \lambda$,
where recall that $\mu > \lambda$.
Then for any event $\mathcal{E}$ which depends on
$\tau$, we can write
\begin{align}
\PP[\mathcal{E}]
&= \PP[\mathcal{E}\,|\,|\tau| \leq \bar{L}] \,\PP[|\tau| \leq \bar{L}]
+ \PP[\mathcal{E}\,|\,|\tau| > \bar{L}]
\,\PP[|\tau| > \bar{L}]\nonumber\\
&\leq \PP[\mathcal{E}\,|\,|\tau| \leq \bar{L}]
+ \PP[|\tau| > \bar{L}]\nonumber\\
&\leq \PP[\mathcal{E}\,|\,|\tau| \leq \bar{L}]
+ \delta_1^2,\label{eq:controllingLength}
\end{align}
for the choice of $\bar{L}$ above.
\paragraph{Events $H^o$}
For $o = -,+$, we use Lemma~\ref{lemma:staying}
to bound the probability of $(H^o)^c$. By construction,
$\tilde{x}^o_1$ is a child of the root, so the edge length between the root and $\tilde{x}^o_1$ is at most $t_{\textnormal{max}}$. Hence, using Lemma~\ref{lemma:staying} and~\eqref{eq:controllingLength} with $\tau := \sigma_\rho$ and $\mathcal{E} := (H^o)^c$,
we get
\begin{equation}
\label{eq:boundH}
\PP[(H^o)^c]
\leq t_{\textnormal{max}} (\bar{L}+1)(\mu + \lambda + \eta)
+ \delta_1^2.
\end{equation}
\paragraph{Events $G^o_k$}
For $o = -,+$ and $k = 1,\ldots,B^o-1$, we use
Lemma~\ref{lemma:staying} together with
Proposition~\ref{prop:Fitch2} to bound
the probability of
$(G^o_k)^c$. Here we take
$\tau := \sigma_{\tilde{x}^o_k}$ and $\mathcal{E} := (G^o)^c$. By assumption, the edge length between $\tilde{x}^o_k$ and its off-backbone child $\tilde{z}^o_k$ is at most $t_{\textnormal{max}}$.
Further, for any fixed failure probability $\delta_a > 0$ and length threshold $\bar{L}$, the maximum branch length $t_{\textnormal{max}}$ can be taken small enough for Proposition~\ref{prop:Fitch2} to hold.
By~\eqref{eq:controllingLength},
we get
\begin{align*}
\PP[(G^o_k)^c]
&\leq \PP[(G^o_k)^c\,|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}] + \delta_1^2\\
&\leq \PP\left[\{\sigma_{\tilde{x}^o_k} \neq \sigma_{\tilde{z}^o_k}\}
\bigcup
\left\{\{\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}\} \cap \{\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}\}\right\}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right] + \delta_1^2\\
&\leq \PP\left[\{\sigma_{\tilde{x}^o_k} \neq \sigma_{\tilde{z}^o_k}\}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right]\\
&\qquad + \PP\left[
\left\{\{\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}\} \cap \{\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}\}\right\}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right] + \delta_1^2
\end{align*}
We use the Markov property to bound
the second term as follows:
\begin{align*}
&\PP\left[
\left\{\{\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}\} \cap \{\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}\}\right\}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right]\\
&= \PP\left[
\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right]
\,\PP\left[
\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}
\,\middle|\,\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}, |\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right]\\
&= \PP\left[
\sigma_{\tilde{x}^o_k} = \sigma_{\tilde{z}^o_k}
\,\middle|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}\right]
\,\PP\left[
\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}
\,\middle|\,|\sigma_{\tilde{z}^o_k}| \leq \bar{L}\right]\\
&\leq \PP\left[
\hat\sigma_{\tilde{z}^o_k} \neq \sigma_{\tilde{z}^o_k}
\,\middle|\,|\sigma_{\tilde{z}^o_k}| \leq \bar{L}\right].
\end{align*}
Plugging this back above and using Lemma~\ref{lemma:staying} and
Proposition~\ref{prop:Fitch2}
gives
\begin{equation}
\label{eq:boundG}
\PP[(G^o_k)^c]
\leq t_{\textnormal{max}} (\bar{L}+1)(\mu + \lambda + \eta)
+ \delta_a + \delta_1^2.
\end{equation}
\paragraph{Events $F^o_k$}
For $o = -,+$ and $k = 1,\ldots,B^o-1$, we use
Lemma~\ref{lemma:atmostone} to bound
the probability of
$(F^o_k)^c$. Here we take
$\tau := \sigma_{\tilde{x}^o_k}$ and $\mathcal{E} := (F_k^o)^c$. By construction (i.e., by the backbone sparsification pre-processing step), the edge length between $\tilde{x}^o_k$ and its backbone child $\tilde{x}^o_{k+1}$ is at most $2 \delta_1$.
By~\eqref{eq:controllingLength}
and Lemma~\ref{lemma:atmostone}, we get
\begin{align}
\PP[(F^o_k)^c]
&\leq \PP[(F^o_k)^c\,|\,|\sigma_{\tilde{x}^o_k}| \leq \bar{L}] + \delta_1^2\nonumber\\
&\leq \left\{2 \delta_1 (\bar{L}+2)[\mu+\lambda+\eta]\right\}^2 + \delta_1^2.\label{eq:boundF}
\end{align}
\paragraph{Union bound}
Taking a union bound over all events above
gives
\begin{align*}
\PP[\mathcal{B}]
&\leq 2 \left[t_{\textnormal{max}} (\bar{L}+1)(\mu + \lambda + \eta)
+ \delta_1^2\right]\\
&\qquad + \sum_{o=-,+}\sum_{k=1}^{B^o-1} \left[
t_{\textnormal{max}} (\bar{L}+1)(\mu + \lambda + \eta)
+ \delta_a + \delta_1^2
\right]\\
&\qquad + \sum_{o=-,+}\sum_{k=1}^{B^o-1} \left[
\left\{2 \delta_1 (\bar{L}+2)[\mu+\lambda+\eta]\right\}^2 + \delta_1^2
\right]
\end{align*}
by \eqref{eq:boundH},~\eqref{eq:boundG} and
\eqref{eq:boundF}.
We make all terms in square brackets of order $\delta_1^2 \log \delta_1^{-1}$ by choosing $\delta_a := \delta_1^2$ and then choosing $0 < t_{\textnormal{max}} \leq \delta_1^2$ small enough
for Proposition~\ref{prop:Fitch2} to hold.
Then
we get, using~\eqref{eq:barLdef},
\begin{align*}
\PP[\mathcal{B}]
&\leq 2 \left[\delta_1^2 ( C' \log(\delta_1^{-1})+1)(\mu + \lambda + \eta)
+ \delta_1^2\right]\\
&\qquad + 2 (B^o-1) \left[
\delta_1^2 ( C' \log(\delta_1^{-1})+1)(\mu + \lambda + \eta)
+ 2\delta_1^2
\right]\\
&\qquad + 2 (B^o-1) \left[
\left\{2 \delta_1 ( C' \log(\delta_1^{-1})+2)[\mu+\lambda+\eta]\right\}^2 + \delta_1^2
\right].
\end{align*}
Because the tree has height $h$ and each backbone edge has length at least $\delta_1$ (after pre-processing), with the exception of the first and last one on each side of the root, we must have $(B^o-2) \delta_1 \leq h$, or after rearranging $B^o \leq h/\delta_1 + 2$. Employing this bound and simplifying gives finally
\begin{equation*}
\PP[\mathcal{B}]
\leq C h \delta_1 \log^2(\delta_1^{-1}),
\end{equation*}
for a constant $C$ depending only on $\mu, \lambda, \eta$,
as claimed.
\end{proof}
\subsection{Proof of the theorem}
We are now ready to finish the proof of the main result.
\begin{proof}[Proof of Theorem~\ref{thm:main}]
For a fixed failure probability $\varepsilon$, we first choose $\delta_1$ small enough (as a function of $h, \mu, \lambda, \eta$) such that $C h \delta_1 \log^2(\delta_1^{-1}) \leq \varepsilon$.
We then choose $t_{\textnormal{max}}$ small enough (again as a function of $h, \mu, \lambda, \eta$)
that Proposition~\ref{prop:Intersection} implies $\PP[\mathcal{B}] \leq \varepsilon$.
Proposition~\ref{prop:correct-align} then completes the proof of the theorem.
\end{proof}
\section*{Acknowledgments}
SR is grateful to Alexandre Bouchard-C\^ot\'e (UBC) for insightful discussions in the early stages of this project.
SR was supported by NSF grants DMS-1614242, DMS-1902892, DMS-1916378
and DMS-2023239 (TRIPODS Phase II), as well as a Simons Fellowship and a Vilas Associates Award. BL was supported by NSF grants DMS-1614242, CCF-1740707 (TRIPODS), DMS-1902892 and a Vilas Associates Award (to SR). BL was also supported by NSF grant DMS-1646108 (University of Michigan-Ann Arbor, Department of Statistics, RTG).
\newpage
\bibliographystyle{alpha}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 2,870
|
Q: Equivariant maps between real linear representations of finitely generated groups Let $G$ be a finitely generated group, ad let $(V,\rho)$ and $(W,\phi)$ be two finite dimensional linear representations of $G$ over $\mathbb{R}$. Given a linear map $f:V\to W$ it's immediate to check whether $f$ is equivariant or not, but I find it hard to find an equivariant map if I'm not given one.
Is there any results that classify or characterize all the equivariant linear maps between the two?
A: Just take an arbitrary linear map and average it over $G$.
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,098
|
Q: how to find the lower limits of integration? Find the volume of the solid obtained by rotating the region bounded by the curves.
$y^2 = x, \ x = 2y; \ about \ the \ y-axis$
Why isn't the lower limit of integration -2?
A: Identify the top and bottom functions:
$$
\begin{align}
\color{red}{y_{t}(x)} &= \color{red}{\frac{1}{2} x}, \\
\color{blue}{y_{b}(x)} &= \color{blue}{x^{2}}.
\end{align}
$$
Bottom intersection at $(0,0)$; top at $\left( \frac{1}{2}, \frac{1}{4} \right)$.
Volume of revolution
$$
V = \pi \int_{0}^{1/2}
\color{red}{\left(\frac{1}{2} x\right)^{2}} -
\color{blue}{\left( x^{2} \right)^{2}} \, dx = \frac{\pi}{240}
$$
|
{
"redpajama_set_name": "RedPajamaStackExchange"
}
| 8,343
|
{"url":"https:\/\/psle-math.com\/student\/test\/Measurement-Questions-Volume","text":"# Question 1 of 50\n\nThe total area of the 6 face of a cube is 150cm$^2$. What is the volume of the cube?\n\nA\n120cm$^3$\nB\n125cm$^3$\nC\n130cm$^3$\nD\n135cm$^3$\nE\nNone of the above","date":"2023-03-24 15:07:58","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.399029403924942, \"perplexity\": 506.58331183497705}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2023-14\/segments\/1679296945287.43\/warc\/CC-MAIN-20230324144746-20230324174746-00129.warc.gz\"}"}
| null | null |
{"url":"https:\/\/en.wikipedia.org\/wiki\/Equivalent_potential_temperature","text":"# Equivalent potential temperature\n\nEquivalent potential temperature, commonly referred to as theta-e ${\\displaystyle \\left(\\theta _{e}\\right)}$, is a quantity related to the stability of a column of air in the atmosphere.\n\n${\\displaystyle \\theta _{e}}$ is the temperature a parcel of air would reach if all the water vapor in the parcel were to condense, releasing its latent heat, and the parcel was brought adiabatically to a standard reference pressure, usually 1000 hPa (1000 mbar) which is roughly equal to atmospheric pressure at sea level. In stable conditions, ${\\displaystyle \\theta _{e}}$ increases with altitude. If ${\\displaystyle \\theta _{e}}$ decreases with height, convection can occur. The comparison of the equivalent potential temperature of parcels of air at different pressures thus provides a measure of the instability of the column of air.[1]\n\n## Explanation\n\n### Stability\n\nCool air is denser than warm air at the same pressure (see gas laws). Like a ball balanced on top of a hill, denser fluid lying above less dense fluid is dynamically unstable: if cool air is positioned above warm air, the former will sink and the latter will rise, the two volumes of air passing around and through each other, and perhaps mixing to some extent, until a stable condition (with denser fluid below and lighter fluid above) is achieved. The temperature near the ceiling of a room is consistently warmer than that near the floor.\n\nIf a hydrostatic fluid is compressible, the criterion for dynamic stability is not simply that denser fluid must lie below light fluid, but that small perturbations must tend to correct themselves. When lower fluid is raised up into upper fluid, (during which process the density of the lower fluid decreases due to the drop in pressure), stability requires that it remain denser than the upper fluid, so that gravity pulls it back toward its original position. The fluid is unstable if small perturbations tend to amplify themselves, i.e. if dense lower fluid, when displaced upward, expands enough to become lighter than the surrounding upper fluid, and therefore continues to move upward.\n\n### Potential temperature\n\nIn the atmosphere, where vertical variation in pressure is much larger than in a room, the situation is complicated by adiabatic temperature change: as a parcel of air moves upward, the ambient pressure drops, causing the parcel to expand. Some of the internal energy of the parcel is used up in doing the work required to expand against the atmospheric pressure, so the temperature of the parcel drops, even though it has not lost any heat. Conversely, a sinking parcel is compressed and becomes warmer even though no heat is added.\n\nAir at the top of a mountain is usually colder than the air in the valley below, but the arrangement is not unstable: if a parcel of air from the valley were somehow lifted up to the top of the mountain, when it arrived it would be even colder than the air already there, due to adiabatic cooling; it would be heavier than the ambient air, and would sink back toward its original position. Similarly, if a parcel of cold mountain-top air were to make the trip down to the valley, it would arrive warmer and lighter than the valley air, and would float back up the mountain.\n\nSo cool air lying on top of warm air can be stable after all (as long as the temperature decrease with height is less than the adiabatic lapse rate); the dynamically important quantity is not the temperature, but the potential temperature\u2014the temperature the air would have if it were brought adiabatically to a reference pressure. The air around the mountain is stable because the air at the top, due to its lower pressure, has a higher potential temperature than the warmer air below.\n\n### Water vapor\n\nA parcel of air containing water vapor, if it rises far enough, cools to its dew point: it becomes saturated with water vapor. This occurs because the vapor pressure of water decreases at lower temperatures (see Clausius\u2013Clapeyron relation). If the parcel of air continues to rise, water vapor begins to condense into liquid droplets. The condensing water releases its latent heat to the surrounding air, partially offsetting the adiabatic cooling. A saturated parcel of air therefore cools less than a dry one would as it rises (its temperature changes with height at the moist adiabatic lapse rate, which is smaller than the dry adiabatic lapse rate).\n\nSaturated air can be unstable even though its potential temperature increases with height: if the warming due to condensation is enough that a parcel of saturated air which is displaced upward (and would otherwise cool to below the ambient temperature) ends up warmer (lighter) than the surrounding air, it will continue to rise. This is the reason for defining the equivalent potential temperature, in analogy with the potential temperature: potential temperature is a temperature adjusted for potential warming due to adiabatic compression; equivalent potential temperature factors in potential warming due to condensation as well. For saturated air, or for air which is likely to be lifted high enough to reach saturation, it is the equivalent potential temperature which must increase with height in order to ensure stability.\n\n## Formula\n\nA number of approximate formulations are used for calculating equivalent potential temperature, since it is not easy to compute integrations along motion of the parcel. Bolton (1980) [2] gives review of such procedures with estimates of error. His best approximation formula is used when accuracy is needed:\n\n${\\displaystyle \\theta _{e}=\\theta _{L}\\exp \\left[\\left({\\frac {3036}{T_{L}}}-1.78\\right)r\\left(1+0.448r\\right)\\right]}$\n${\\displaystyle \\theta _{L}=T\\left({\\frac {p_{0}}{p-e}}\\right)^{\\kappa _{d}}\\left({\\frac {T}{T_{L}}}\\right)^{0.28r}}$\n${\\displaystyle T_{L}={\\frac {1}{{\\frac {1}{T_{d}-56}}+{\\frac {\\log _{e}(T\/T_{d})}{800}}}}+56}$\n\nWhere:\n\n\u2022 ${\\displaystyle \\theta _{L}}$ is (dry) potential temperature [K] at the lifted condensation level (LCL),\n\u2022 ${\\displaystyle T_{L}}$ is (approximated) temperature [K] at LCL,\n\u2022 ${\\displaystyle T}$ is temperature [K] of air at pressure ${\\displaystyle p}$,\n\u2022 ${\\displaystyle T_{d}}$ is dew point temperature at pressure ${\\displaystyle p}$,\n\u2022 ${\\displaystyle p}$ is pressure at the point [hPa or mbar],\n\u2022 ${\\displaystyle p_{0}}$ is standard reference pressure (1000 hPa),\n\u2022 ${\\displaystyle e}$ is the water wapor pressure (to obtain ${\\displaystyle \\theta _{L}}$ for dry air),\n\u2022 ${\\displaystyle \\kappa _{d}=R_{d}\/c_{pd}}$ is the ratio of the specific gas constant to the specific heat of dry air at constant pressure (0.2854),\n\u2022 ${\\displaystyle r}$ is mixing ratio of water vapor mass per mass [kg\/kg] (sometimes value is given in [g\/kg][3] and that should be divided by 1000).\n\nA little more theoretical formula is commonly used in literature like Holton (1972) [4] when theoretical explanation is important:\n\n${\\displaystyle \\theta _{e}\\approx \\theta _{L}\\exp \\left[{\\frac {r_{s}(T_{L})L_{v}(T_{L})}{c_{pd}T_{L}}}\\right]}$\n\nWhere:\n\n\u2022 ${\\displaystyle r_{s}(T_{L})}$ is saturated mixing ratio of water at temperature ${\\displaystyle T_{L}}$ which is approximately considered same to specific humidity in low temperature,\n\u2022 ${\\displaystyle L_{v}(T_{L})}$ is latent heat of evaporation at temperature ${\\displaystyle T_{L}}$ (2406 kJ\/kg {at 40\u00a0\u00b0C} to 2501 kJ\/kg {at 0\u00a0\u00b0C}), and\n\u2022 ${\\displaystyle c_{pd}}$ is specific heat of dry air at constant pressure (1005.7 J\/(kg\u00b7K)).\n\nFurther more simplified formula is used (in, for example, Stull 1988[5] \u00a713.1 p.\u00a0546) for simplicity, if it is desirable to avoid computing ${\\displaystyle T_{L}}$:\n\n${\\displaystyle \\theta _{e}=T_{e}\\left({\\frac {p_{0}}{p}}\\right)^{\\frac {R_{d}}{c_{pd}}}\\approx \\left(T+{\\frac {L_{v}}{c_{pd}}}r\\right)\\left({\\frac {p_{0}}{p}}\\right)^{\\frac {R_{d}}{c_{pd}}}}$\n\nWhere:\n\n\u2022 ${\\displaystyle T_{e}}$ = equivalent temperature\n\u2022 ${\\displaystyle R_{d}}$ = specific gas constant for air (287.04 J\/(kg\u00b7K))\n\n## Usage\n\nBack trajectories of air masses between December 31, 1997, and January 1998 which caused the North American Ice Storm of 1998\n\nThis applies on the synoptic scale for characterisation of air masses. For instance, in a study of the North American Ice Storm of 1998, professors Gyakum (McGill University, Montreal) and Roebber (Wisconsin University, Milwaukee) have demonstrated that the air masses involved originated from high Arctic at an altitude of 300 to 400\u00a0hPa the previous week, went down toward the surface as they moved to the Tropics, then moved back up along the Mississippi Valley toward the St. Lawrence Valley. The back trajectories were evaluated using the constant equivalent potential temperatures.[6]\n\nIn the mesoscale, equivalent potential temperature is also a useful measure of the static stability of the unsaturated atmosphere. Under normal, stably stratified conditions, the potential temperature increases with height,\n\n${\\displaystyle {\\frac {\\partial \\theta _{e}}{\\partial z}}>0}$\n\nand vertical motions are suppressed. If the equivalent potential temperature decreases with height,\n\n${\\displaystyle {\\frac {\\partial \\theta _{e}}{\\partial z}}<0}$\n\nthe atmosphere is unstable to vertical motions, and convection is likely. Situations in which the equivalent potential temperature decreases with height, indicating instability in saturated air, are quite common.","date":"2016-12-07 22:45:45","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 31, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.6508265137672424, \"perplexity\": 967.5791907617955}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2016-50\/segments\/1480698542250.48\/warc\/CC-MAIN-20161202170902-00262-ip-10-31-129-80.ec2.internal.warc.gz\"}"}
| null | null |
Lindemann est un nom propre germanique qui peut faire référence à :
Patronyme
Adzire Lindemann (1882-1945), résistante française ;
Andreas Lindemann (1943-), théologien et bibliste allemand ;
Ernst Lindemann (1899-1941), officier de marine allemand, commandant du cuirassé allemand Bismarck ;
Ferdinand von Lindemann (1852-1939), mathématicien allemand ;
Théorème d'Hermite-Lindemann ;
Théorème de Lindemann-Weierstrass ;
Frederick Lindemann (1886-1957), physicien anglais et proche de Winston Churchill ;
Fritz Lindemann (1894-1944), General der Artillerie allemand pendant la Seconde Guerre mondiale ;
Georg Lindemann (1884-1963), Generaloberst allemand pendant la Seconde Guerre mondiale ;
Gerhard Lindemann (1896-1994), Generalmajor allemand pendant la Seconde Guerre mondiale ;
Guido Lindemann (1955-), joueur de hockey sur glace suisse ;
Kim Lindemann (1982-), joueur de hockey sur glace suisse ;
Laura Lindemann (1996-), triathlète allemande ;
Maggie Lindemann (1998-), chanteuse américaine ;
Markus Lindemann (1953-), joueur de hockey sur glace suisse ;
Max Lindemann, entrepreneur français, fondateur de la marque Etam ;
Stefan Lindemann (1980-), patineur artistique allemand ;
Sven Lindemann (1978-), joueur de hockey sur glace suisse ;
Till Lindemann (1963-), chanteur, poète et acteur allemand, membre du groupe Rammstein et du groupe Lindemann.
Musique
Lindemann, groupe germano-suédois de metal industriel fondé en 2015.
Patronyme germanique
Patronyme suédois
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 3,284
|
Hartzell Announces New Co-Presidents
Date: January 31, 2000 Category: Press Releases
Next Generation of Brown Family Takes Over Daily Operations
PIQUA, Ohio – January 31, 2000 — Hartzell Propeller chairman James W. Brown, Jr. recently announced the promotions of Joseph W. Brown and James W. Brown III to the positions of Co-Presidents for the company. In their new capacities, Joseph W. Brown serves as co-president of operations and James W. Brown III serves as co-president of engineering and customer services.
"I'm not retiring, but I am looking forward to enjoying some longer weekends," said Hartzell chairman James W. Brown, Jr. "I'm going to focus on strategic planning issues while Joe and Jim take over the day-to-day operations…the company will do as well or better under this new organization."
Under the new structure, Joe Brown oversees manufacturing, purchasing, quality assurance and information technology, while Jim Brown III oversees marketing, sales, product support, engineering, distribution and the Hartzell Service Center.
Joseph W. Brown joined Hartzell Propeller in 1990 and served as assistant to the manufacturing vice president. He was promoted to the position of production control manager in 1992 where he developed production schedules and matched plant capacity with market demand. In 1994 he assumed responsibility for manufacturing and was named director of manufacturing. Brown became the vice president of manufacturing in 1995 where his duties included production control, manufacturing, manufacturing engineering and shipping.
James W. Brown III joined Hartzell Propeller in 1993 as the operations supervisor for its Service Center where he had day-to-day responsibility for all production aspects of the factory-owned repair station. A year later, Brown became the director of the Service Center and in 1995 he was promoted to the position of vice president of parts and service. His responsibilities included the oversight and management of the FAA-approved manufacturer's maintenance facility and the factory repair station which provide parts and service to Hartzell's aftermarket customers. Brown was also responsible for Hartzell's international distributor network.
Hartzell Propeller Inc. is the world's leading manufacturer of propeller systems, well known for its advanced engineering and manufacturing capabilities. The company is rich in aviation heritage tracing its beginnings to relationships with Orville Wright and Glenn Curtiss. Significant technical innovations include the industry's first full-feathering propeller for light twins, the first fully reversing propellers for corporate turboprops and the industry's first composite structure blades. Hartzell was recently selected by NASA as the exclusive propeller partner for its General Aviation Propulsion (GAP) program-developing the next generation of general aviation propulsion systems. For more information on Hartzell, visit the Web at www.hartzellprop.com.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 4,738
|
{"url":"https:\/\/gamedev.stackexchange.com\/questions\/49686\/java2d-collision-detection?noredirect=1","text":"# Java2D Collision Detection\n\nI've been wrapping my head around Java2D collision detection. I am working with Slick library and trying to figure a simple clean code to check if a collision has occurred.\n\nThe idea is fairly simple, I have an Entity class seen as a bound box (a rectangle in my case,but I am not using the intersects method), I am trying to run a check if current entity is colliding with any other entities passed in an arraylist:\n\n public boolean collides(Object o){\nEntity e = (Entity) o;\nreturn x + width > e.x && x < e.x + e.width && y + height > e.y && y < e.y + e.height;\n}\n\npublic boolean collisidesAny(ArrayList al){\nfor(Object o: al){\nreturn collides(o);\n}\nreturn false;\n}\n\n\nNow the thing with this code is that it gives me quite a bit of headache. The second method will return true the moment the objects do collide, and if I was to use it with a keyboard I wouldn't be able to move back since it won't allow any commands. Naturally I tried with sending the entity one step back when collision does occur, but the problem is I can't know where did it occur from, so I would have to do a step back both x and y wise. Other way I tried was a priori check, where I would check if a collision would occur if a step was made and if it wasn't I would allow the step, but this has gotten really complicated and buggy very quickly. Therefor I would like to know if there was a way to do collision this way?","date":"2020-09-27 16:20:23","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 1, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.30365732312202454, \"perplexity\": 849.1306129629322}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2020-40\/segments\/1600400283990.75\/warc\/CC-MAIN-20200927152349-20200927182349-00657.warc.gz\"}"}
| null | null |
Todas as cordas
Para Baixo
Todos os
Accessórios
Em Busca
do Timbre
das Cordas
A CHORD
Slash Expressando
sem Palavras
0 Itens no seu pedido
Older Episode
"We all looked for that perfect gig ya know?.. I've had a constant career and I've influenced a lot of people and I've been making great music. I'm too old for regrets now."
View Other Episodes
Walter Schreifels
Frank Iero
Tyler Bryant
The War on Drugs - Adam Granduciel
Andy McKee
Bassel Hallak
Best Coast - Bobb Bruno
Butch Walker
Act of Defiance - Chris Broderick
Chris Vega
Christian Adameit
Zac Brown Band - Clay Cook
System Of A Down - Daron Malakian
Jane's Addiction - Dave Navarro
thenewno2, Fistful of Mercy - Dhani Harrison
Thrice - Dustin Kensrue
My Chemical Romance - Frank Iero
Hanne Kah
Hannes Kelch
Nine Inch Nails, Angels & Airwaves, The New Regime - Ilan Rubin
Dinosaur Jr. - J Mascis
Jacky Bastek
AFI - Jade Puget
Alice in Chains - Jerry Cantrell
Eagles of Death Metal - Jesse Hughes
Jimmy Eat World - Jim Adkins
Rascal Flatts - Joe Don Rooney
Dream Theater - John Myung
Dream Theater - John Petrucci
The Smiths - Johnny Marr
Phantogram - Josh Carter
TOOL - Justin Chancellor
Kai Stuffel
Kenny Wayne Shepherd Band - Kenny Wayne Shepherd
Metallica - Kirk Hammett
Against Me! - Laura Jane Grace
Deap Vally - Lindsey Troy
Halestorm - Lzzy Hale & Joe Hottinger
Mötley Crüe - Mick Mars
Green Day - Mike Dirnt
MxPx - Mike Herrera
Journey - Neal Schon
311 - Nick Hexum and Tim Mahoney
Racer X, Mr. Big - Paul Gilbert
KISS - Paul Stanley
Of Mice & Men - Phil Manansala & Alan Ashby
Larkin Poe - Rebecca Lovell & Megan Lovell
Pantera, Down, Kill Devil Hill - Rex Brown
Nine Inch Nails - Robin Finck
Seether - Shaun Morgan
System Of A Down - Shavo Odadjian
Toto - Steve Lukather
Billy Idol - Steve Stevens
Avenged Sevenfold - Synyster Gates & Zacky Vengeance
Thomas Blug
Rise Against - Tim McIlrath
No Doubt, Dreamcar - Tom Dumont
Albert Lee:
I used to listen to the radio a lot when I was a kid. I'd hear Doris Day and Guy Mitchell and all those things. I guess I had an interest in music at that time.
I think Christmas of 1958, I was 15. It was a Hofner archtop guitar. I actually played that for about a year, year and a half maybe. I was really making good progress. I'd become a big fan of Gene Vincent and his guitar player, Cliff Gallup, and Buddy Holly. I'd seen The Chirping Crickets album, Buddy Holly's first album, and he had this guitar with three pickups and a weird kind of tremolo on it. I thought, "Wow, that's amazing." And I was in the West End of London, there was a music store there, and they had a guitar that looked very much like it in the window.
So I'd go into [Selma's 00:01:35] on a Saturday, go up there and look at the guitars and say, "Well, what would suit me best?" I didn't know at that time. Lo and behold, this guy heard me playing in the store. He said, "Oh, you're good. Are you in a band?" I said, "Well, kind of in between. Why?" He said, "well, we're looking for a guitar player. You won't need a guitar. We've got one already, you can use." So I went over to his house and it wasn't too far away. I opened up this guitar case. And there's a Les Paul custom, with a bigsby. I thought, "wow, okay, I'll join your band."
I was listening to the rock and roll that had come in. Bill Haley was really big in '56, '57 in England, but soon after that we had Buddy Holly and Gene Vincent and The Everly Brothers and Elvis, of course. I was a big fan of all these guys and I just loved the sounds of the guitars and tried to copy a lot of the solos. And I did quite well on some of the simpler ones.
I was playing in the 2i's Coffee Bar. It was a coffee bar, right in the west end of London. I became part of the house band there. It was just three of us. People would come in and get up and sing songs, and just play things on the fly, and no rehearsals or anything. It was a good time. And a lot of musicians came through that club. When I used to play there, Jimmy Page used to come in and we became good friends and went to his house and he came to mine and we'd play records. He was a big James Burton fan and I was more into Scotty Moore. By the time I was playing at the 2i's Coffee Bar, I realized I was earning more money playing the guitar than I could paint spraying or whatever.
I've got kind of a reputation around town. I mean, there were some good players around. I didn't really know Jeff Beck at that time, but Jimmy went on to play with the Yardbirds with him. I joined Chris Farlowe. That was a good step forward. I played at lots of clubs and we were based at a jazz R&B club in the center of London called The Flamingo.
I took the pickup cover off of the rhythm pickup of my Telecaster and I broke the windings. And I thought, "that wasn't a smart thing to do." So I thought, "what am I going to do now?" You couldn't walk into a store and buy a pickup in those days, in the early '60s. I took the middle pickup out of this SG and put it on the Tele. I liked to think I was one of the first people to do that. I remember doing a tour with Chris Farlowe in the mid '60s. We were on a show with the Paul Butterfield band, and Mike Bloomfield saw my tele with the humbucker on it and he said, "boy, that's a great idea, I've never seen that before."
I found out that there were some bands around London playing in pubs that were playing country music. And the one night I decided to go, was this really good band and a good guitar player who also played steel guitar. So I used to make a regular trip to this pub when I wasn't working myself and sit in with these guys. And I thought, "whoa, this is great. I love this." The singer with the band said, well, maybe we should put together a little band, which we did. It was a four piece and we called ourselves Country Fever. I got that name from a title of a Ricky Nelson album he did, called Country Fever. A lot of the people in the pubs would say, "oh yeah, you guys are good, but you're not as good as the Americans." And I knew we were.
Well, actually, I should back up a bit because you'd like to know about Ernie Ball, of course. So when I was playing with, with Heads Hands & Feet, Ernie heard our Country Boy record on the radio. It was getting played a lot. So he and Sterling came to a couple of gigs and we got to be good friends. And they said, "oh, you should come down to the factory," which was in Newport at the time. What I used to do is buy a set of strings, it was an expensive way of doing it, buy a set of strings, throw away the bottom strings, move them all down one and then put on a really light first string. And then I'd have a rock and roll set. But then of course I discovered the Ernie Ball strings and Sterling would have a go at me, "oh, when you're going to start you using our strings?" So I did eventually. And that was how long ago? 50 years ago, maybe almost.
So I did this gig with the Crickets. At the end of the gig, they said, "oh, that was great. You want to do the whole tour?" I said, "okay, love to." I was still hanging out with the Heads Hands & Feet guy. Our manager at the time was managing Joe Cocker. Joe had supposedly been rehearsing for this major tour. It was meant to be his first big tour after Mad Dogs and Englishmen. Some of the guys had a big fight and they needed a guitar player and drummer immediately. And the tour was about to start in two weeks time. They were headliners around the US. I joined the band with the drummer from Heads Hands & Feet, and off we went on the road with Joe.
I flew back to England to work on a record and who should be on the session, but Eric and half his band.
Speaker 3:
For video purposes, Eric who?
Oh, Clapton.
Just in case.
Is there any other? [crosstalk 00:09:42].
So there was in England recording this album with Eric and Marc Benno, and at the end of the sessions, Eric's manager came up to me and said, "Eric did a tour without a rhythm guitar player, without a second guitar. Would you be interested in going out with Eric?" I thought, "wow. Yeah. Why not?" That was early '79. So off I went on the road with Eric. At that first rehearsal, that's when he gave me that Les Paul. He was a lot of fun in those days and he still is fun, but in a different way now.
I got this call from a guy named Terry Slater who worked with the Everly Brothers, and he said, "well, looks like the guys are going to give it another shot and they're going to do this reunion concert at the Royal Albert Hall in London. Are you available?" And I said, "well, yes. Great." And apparently, they both agreed that I should be one of the guitar players, which made me feel really good. And I thought, "well, this is great. I'll get to play with the Everly Brothers." In the back of my mind, I thought, "well, how long is going to last?" Because it was a pretty volatile breakup that they had. To everyone's surprise, it went on for 26 years.
We all look for that perfect gig. It's the way I've looked at it. It makes me feel better, not having a million dollars. At least I've had a constant career and I've influenced a lot of people and I've been making great music. I'm too old for regrets now. I could have done things a little differently in the '60s. I should have done... But the guitar players that I knew back then were focused on the blues, like Eric and Jeff Beck. That was the time that I got waylaid by country music. So I still like what they were doing and that kind of music, but I was still influenced by the country side of it.
My style kind of covers a lot of ground. I'm influenced by country and I use my fingers a lot. I find that I can play most things, but I like to think that I've mastered a really nice technique over the years. I've always been adaptable. I could play along with Eric. In fact, he gave me lots of solos in those days. When I was playing with, with Joe Cocker and Bill Wyman and Chris Farlowe, I was playing a lot of R&B. I like to think that my playing is melodic and tasteful and it seems to fit in with most things I'm asked to do.
© 2022 Ernie Ball Inc.
Assine e tenha um passe livre para ser o primeiro a saber sobre novos produtos, concursos, e muito mais.
Eu toco: Guitarra Baixo Violão
Cadastre-se no mailing da Ernie Ball. Você pode se descadastrar a qualquer momento.
Junte-se ao legado
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 8,696
|
{"url":"http:\/\/math.stackexchange.com\/tags\/hyperbolic-geometry\/hot","text":"# Tag Info\n\n5\n\nYes, this is indeed a form of duality. The following outline follows the approach from Perspectives on Projective Geometry by J. Richter-Gebert. Cayley-Klein metric Probably the best way to understand this is probably using Cayley-Klein metrics. That's a very general way to measure distances and lengths in projective geometry. You start by fixing one ...\n\n4\n\nThe usefulness of the Poincar\u00e9 model (or of Klein's) is that it shows that if Euclidean geometry is consistent, then also hyperbolic geometry is. This is because the hyperbolic axioms are true in the model, so a contradiction in hyperbolic geometry would yield a contradiction in Euclidean geometry. It would be wrong to prove theorems in hyperbolic geometry ...\n\n4\n\nNo. There is an infinite-dimensional space of homeomorphisms of the circle; the space of Mobius transformations is the 3-dimensional $PSL_2(\\Bbb R)$. But more pointedly, the restriction of a Mobius transformation to the ideal boundary can have at most two fixed points if it's not the identity; but any closed subset of $S^1$ is the set of fixed points of some ...\n\n4\n\nAs proposed in the comments, I posted this question on MathOverflow and it got an answer there. Here is the link: The question on MathOverflow (answered) Thanks to everyone who read the question and thought about it.\n\n3\n\nThat is not true. A simple example is the following one: consider the hyperbolic isometry \\begin{equation*} \\phi:= \\left[ \\begin{array}{l l} 4 &0\\\\ 0 &\\frac{1}{4}\\\\ \\end{array} \\right] \\in\\text{PSL}(2,\\mathbb{R})=\\text{Isom}^+\\left(\\mathbb{H}^2\\right) \\end{equation*} and the finite order elliptic element \\begin{equation*} \\rho:= \\left[ \\begin{...\n\n2\n\n\nOnly top voted, non community-wiki answers of a minimum length are eligible","date":"2016-07-27 06:14:47","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 1, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.9937708973884583, \"perplexity\": 343.5081071926607}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2016-30\/segments\/1469257825366.39\/warc\/CC-MAIN-20160723071025-00094-ip-10-185-27-174.ec2.internal.warc.gz\"}"}
| null | null |
Buster's Notes
The Piles
My Life in Weeks
Codex Vitae
Game the system
Exploiting a game's design flaws.
The original post is here: https://erikbenson.typepad.com/mu/2004/0...
There must be a better phrase for this, "gaming the system" sounds really lame. Actually, just the word game makes me feel a bit icky–and that's multiplied by ten when used as a verb. But lately I've been seeing just how much we game the system, any system, all systems, all the time.
To game the system, as I understand it, is to ignore the interface that the system has given you, to ignore the "best" choice (where best is defined by core values and pure motives–before you game it), and to instead try to fight against an outcome that you foresee happening before it has actually happened. You are second-guessing an undesirable outcome (perhaps one where you think you would lose if you were to play completely along the implicit rules of the system) and trying to outsmart it, even as your opponent (maybe this is how it's game-like) tries to second-guess and outsmart you.
Who are you endorsing for the primaries? How much game is involved? How do you reconcile the difference between who you feel the "best" candidate is versus the one who you think has the "best" chance of winning? It is a double game because in this game called the election the "best" chance is the one that everyone votes for after they've decided who the best chance is. It's a self-fulfilling game where you have to take into account how everyone else is going to game the system and try to game it even more. The only problem that really bothers me is that we're always trying to second guess Middle America–you know, where the dumb people live. Certainly they won't choose the best candidate in your mind, and so you have to choose someone else because you're pretty sure there are more dumb people than smart, and it's better to vote slightly dumber if it means you win than to vote smart if it means you'll lose. By trying to outsmart the dumb people you'll end up downgrading the intelligence of your own decision by even more than necessary because we always overestimate just how dumb other people are. They're second-guessing you too… and have more power because they don't feel like they have to compromise their vote to game you. They just play dumb. And that's just what the smart people want you to do. They want you to act dumb. Because then you are. That's the game. I'm starting to like the word game.
Work is a system that we game. How many people would go to work if they didn't get paid? These people are gaming the system–acting in ways that they do not consider to be the best ways in order to get a result that the system is willing to give them if they pretend (very loosely) that they really would go to work even if they didn't get paid. The best game players are those that can convince even themselves. I've pretty much convinced myself that I enjoy my job, and yet I still wish I would've had the day off. Who's gaming who?
Non-zero sum systems are games. I pay you more than something is worth in order to get something that is cheaper than I would've had to get it elsewhere. Both parties win, the sum is greater than its previous values. Out of 3 dollars and 37 cents comes the enjoyment of making $3 profit off of the cost of a mocha and the enjoyment of a sugar and caffeine high at a highly stylized coffee house. It's gaming the system because both parties feel that they're getting the better deal, and giving something to the other person that they don't value as highly as the other. I'm pretty sure this is crossing the line between gaming the system and is now entirely in the territory of something completely different. I suppose it's not gaming unless you're being a little sneaky about it. But most of the time people know when other people are gaming you… they just decide to let it slide so they can be sneaky back. A mutual contract to be sneaky in order to make everyone happy. Come to think of it, this is an essential component to doing things that you wouldn't normally do (because you'd feel guilty). If both people are being sneaky, and you know that for this reason the other person won't expose your sneaky behavior, it's like it doesn't exist. Except to G-O-D. Shh!
Calling in sick when you aren't really sick because you have a pre-allotted number of paid sick days. Jaywalking. Speeding. Wearing the same shirt two days in a row because you'll be seeing different people and don't feel like doing the wash.
The thing about gaming the system is that it is so effective. It is powerful simply because of its effectiveness–it's more effective than appropriate behavior would have been. You think, at least. That is why we decide to game in the first place.
We ask people what they think it means when they rate an item on Amazon. There is what they think it should mean, and then there's what they think Amazon does with the rating. In the end, they'll end up rating things that they don't know anything about even though they know they shouldn't if they were taking the action on a completely semantic level. But because it's a part of a system, and that system takes their input in a way that has some side effects, they can second-guess those side effects and actually produce them intentionally when desired.
Your grandma asks you how you liked that sweater. You say one thing. Your friends ask the same, a little later. You say another thing. Is this still gaming the system? Being nice when you aren't feeling nice. Sharing when you don't want to share. Lying to save feelings. Entertaining when you don't feel like entertaining. Complying with all the pressures of society so that you don't stand out too much (some slips by).
This is a theme lately. Might as well stomp it to death.
January 19, 2004 · In these piles: game-theory · Original post
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 6,053
|
# Table of Contents
Skill Up: A Software Developer's Guide to Life and Career
Credits
About the Author
www.PacktPub.com
eBooks, discount offers, and more
Why subscribe?
Customer Feedback
Preface
What this book covers
Who this book is for
Conventions
Reader feedback
Customer support
Errata
Piracy
Questions
I. Coder Skills
1. Discovering the Tipping Point for Developers
Tipping point for developers
My own experience
The doubt machine
The painful process
The tipping point(s)
The first tipping point
The second tipping point
The secret
The book
The solution
2. Are Developers Born or Made? – Debunking the Myth of Prodigies
Are prodigies real?
The Mozart case study
Are developers born or made?
The tipping point
Why we love the prodigy myth
3. Do You Have to Be a Genius to Be a Developer?
The running man
Do you have to be a genius to be a developer?
The way the mind works
The reason
A smarter approach
4. How to Study and Understand Complex Topics?
A system for how to study
5. Effective Study Practices for Developers
Why traditional study habits don't work
An effective study practices case study
The reification example
The hard way
Additional negative effects
The comprehensive study system
Summary
6. Defining Deep Work and What It Means for Developers
Definition of deep work
The deep work strategy for developers
Taking action
Removing distractions
Study hard and smart
Multiple sessions
Summary
7. Task Switching Costs for Developers
A system for decreasing task switching costs
8. How to Use Willpower Limits Instead of Letting Them Use You?
What are willpower limits?
How many decisions do you make each day?
Why is willpower important?
Are willpower limits real?
When the willpower well runs dry
Saving up willpower
One outfit to rule them all
Being a copycat
Focusing willpower
Summary
9. Cramming Versus Consistent Study and a Study System that Works
10. Is Reading Important for Developers?
Why is reading important for developers?
CEOs and reading
Compounded learning
A compounded learning case study
The CEO who didn't have time to read
My reading system
The reading schedule
Audio books are books too!
Books are too expensive
Summary
11. Learning How to Code – Getting Past Skill Plateaus
What is a learning plateau?
False ceiling
Getting past skill plateaus
Proper information/resources
Best practices
Challenging/new tasks
Frustration = skill
Summary
12. Developer Learning Curve – Why Learning How to Code Takes So Long
What is the learning curve?
The developer learning curve
Liftoff
The twilight zone
The zone
A unique journey
Summary
13. Slowing Down to Learn How to Code Faster
Learn how to code faster
Our default mind
Hacking the mind
Slowing it down
Bend it like Beethoven
From classical music to coding
A practical system
14. Mental Models for Learning How to Code and Improve as a Developer
Mental models for the Kouros
What are mental models?
Mental models for developers
Summary
15. A Developer's Guide for Hacking Procrastination to Achieve Success
Root causes of procrastination
Hacking procrastination
Hacking perfectionism
Hacking the fear of success
Hacking the plan
Summary
16. The Problem with Procrastination for Developers
The problem with procrastination
Instant gratification
Baby steps to knock out procrastination
Baby coding steps
17. Practical Ways to Use the Pomodoro Technique as a Developer
Practical ways to use the Pomodoro Technique
Taking a break
Lifestyle versus fads
A lifestyle of productivity
Practical implementation
18. The Power of Making Mistakes – Learning by Failing
The secret weapon to mastery – making mistakes
Making mistakes – memory steroids
Mistakes force learning
Mistakes kill pride
Summary
19. Learn How to Code – The Guide to Memorization
The guide to memorization
Repetition
Smarter, not harder
Visual mental mapping
Short-term versus long-term memory
Implementing visual mental mapping
Taking a real-world example
Finding patterns
Copy and paste is the enemy
Not everything has to be memorized
20. A System for Learning a New Programming Language
21. Development Study Tips – Reverse Note-Taking
The problem with traditional note-taking
Reverse note-taking
Benefits of reverse note-taking
Narrowed focus
Story-based mindset
Forced repetition
Summary
II. Freelancer Skills
22. Tips for Organically Growing a Freelance Business
Organically growing a freelance business
Referral requests
Blogging
Expert positioning
Open source contribution
Social media marketing
Summary
23. Freelancing Tips – Knowing When to Fire a Client
My urgent client
When to fire a client
#1 – being treated like an employee
#2 – tyranny of urgent
#3 – toxic environment
The joy of firing a client
24. Dodging Silver Bullets for Scalable Freelance Projects
The problem with silver bullets
Silver bullet customization
Becoming a sharp shooter with code libraries
25. A Freelance Guide to Managing Advanced Features
Managing advanced features
The talent pool
The process
Kanban
The result
Summary
A caveat
26. Freelancer Interviews – Practical Tips for Taking Over a Legacy Application
27. Five Tips for Taking Over a Legacy Application
Tips for taking over a legacy application
Creating a test suite
Adding new features via TDD
Breaking out specific features into microservices
DRY up the codebase
Summary
28. Guide to Freelancing – Starting Over Versus Refactoring
The legacy scenario
Starting over versus refactoring
#1 – removing the fear factor
#2 – analyzing the 80/20 principle
#3 – building an automated bug list
#4 – becoming the client
When should you start over?
Summary
29. Should You Use TDD on Freelance Projects? – Comparing Quality Versus Speed
Quality versus Speed
TDD on freelance projects
Making the decision
Giving no choice
Letting the client decide
Using common sense
30. Automating Client Updates as a Freelance Developer
Importance of daily updates
An example of client update
Automating client updates
Version control to the rescue
Summary
31. Freelance Requirement Elicitation – A Guide for Feature Development
Freelance requirement elicitation
How it started
The build
The problem
Who was at fault?
A better way
Step 1
Step 2
A better ending
Summary
32. How to Remotely Demo Work for Freelance Clients?
Why proper demonstrations are important
Review of services to remotely demo work
Screencast
A remote desktop
PowerPoint
Summary
33. Defining Project Success as a Freelance Developer
A clear end
What is scope creep?
When scope creep isn't scope creep
When scope creep goes badly
Based on requirements
Based on a story
The sign off
Summary
34. Top Project Management Tools for Freelancers
Top project management tools
Basecamp
Trello
LeanKit
ProWorkflow
Wrike
GitHub
Summary
35. Top Freelance Bookkeeping Options for Developers
Freelance bookkeeping options
FreshBooks
How it works
FreshBooks additional features
Weaknesses
QuickBooks
NetSuite
Summary
36. Learning the Secret to Get New Clients as a Freelancer
Where to find new clients
The challenge in getting new clients with outsourcing services
Getting new clients as a freelancer
Proposal material
Sending out constant proposals
The result
Summary
37. Managing Client Conflicts as a Freelancer
Strategies for managing client conflicts
38. Examples of Freelance Portfolios That Help Acquire New Clients
Examples of freelance portfolios
Social network utility
An API tool
An accounting application
A scheduling application
A frontend application
39. Importance of Test-Driven Development for Coders
Importance of test-driven development
Summary
40. SEO Best Practices and Strategies for Freelancers
SEO best practices tutorial
Content is king
Creating an XML sitemap
Mixing text, images, and videos
Managing your site speed
Site responsiveness
Backlinks
Focused content
Summary
41. Client Communication Freelancing Tips
A system to maintain proper client communication
Summary
42. Outsource Web Developers Properly with System-Based Processes
A system to manage outsourced web developers
Summary
43. How to Create Accurate Freelance Bids?
Summary
44. Freelancer Tips – Three Ways to Get New Clients
Freelancing services
LinkedIn
Referrals
Summary
III. Career Skills
45. Should I Learn to Code? – A Balanced Perspective on Programming
Should I learn to code? – a balanced look at both sides
Summary
46. Following Your Passion – Good or Bad Advice for Developers?
Following your passion – a case study
Summary
47. How to Learn to Code from Scratch? – A Practical Strategy
Small bites
Tutorials
Reading
Real-world projects
Coding is hard
But you can learn programming
48. How to Choose a Developer Specialty?
How to choose a developer specialty?
#1 – the full stack developer
#2 – the server-side developer
#3 – the frontend developer
#4 – the mobile developer
#5 – the data scientist
Making the decision
49. How to Choose Your Next Programming Language?
How to pick a programming language?
The next job you want
Your specialty
Specialty-based mapping
Summary
50. Developer Soft Skills – Learning How to Gain an Edge in the Marketplace
Developer soft skills
Writing
Conversation
Conversation tips
Management
Design
Public speaking
Becoming a better public speaker
The importance of soft skills
51. Developer Learning Options – A Practical Analysis
Degrees of programming expertise
Becoming a professional developer
Developer bootcamps
Is this practical?
Improving your skill in your current profession
Is this practical?
Learning for fun or as a hobby
Summary
52. Is it Possible to Lose Your Coding Skills?
Summary
53. Is Writing Bad Code Immoral for Developers?
How to write better code
Summary
54. Inspirational Programming Advice from Howard Roark
55. Best Practices Versus Creativity as a Developer
56. A Practical Guide to Approaching Project Development
Student question
Strategies to approaching project development
Planning a feature from start to end
Moving from requirements to stories
Starting with a base case
Fear of the unknown
Moving fast and breaking things
Battling procrastination
Small, practical steps
Getting unstuck
Application bugs
Messages over models
57. How to Practice Programming Techniques and Improve as a Developer?
Engaging in pair programming
Utilizing open source software
Visiting the DailyProgrammer subreddit on Reddit
Taking online courses
Code katas
Summary
58. What Does It Take to Become a Great Developer?
Tips for becoming a great developer
Working through difficult features
Community contribution
Artistry
Craftsmanship
Steve Jobs's craftsmanship
Adapting to change
Tireless learning
Summary
59. How to Stay Sharp as a Developer?
Tips to stay sharp as a developer
#1 – coding exercises
Example coding exercises
#2 – teaching others to code
How does this apply to development?
#3 – reading
#4 – newsletters
#5 – tutorials
Summary
60. Developer Resume Tips – How to Create an Effective Resume?
Developer resume tips
Keep it simple
Keep it relatable
Keep it professional
Summary
61. Developer Salary Negotiation Strategies
Knowing your skill set
Knowing the industry
Knowing the organization
Researching salary rates
62. Best Questions to Ask During a Job Interview
Best questions to ask during a job interview
Poor questions to ask during a job interview
Summary
63. Answering in an Impossible Interview
Questions
Answering impossible interview questions – case studies
64. Greatest Weakness Answers for Coding Interviews
Bad answers to your greatest weakness
Good answers to your greatest weakness
65. Enterprise Software Job Strategy and Guide
Summary
Index
# **Skill Up: A Software Developer's Guide to Life and Career**
* * *
# Skill Up: A Software Developer's Guide to Life and Career
Copyright © 2017 Packt Publishing
All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, without the prior written permission of the publisher, except in the case of brief quotations embedded in critical articles or reviews.
Every effort has been made in the preparation of this book to ensure the accuracy of the information presented. However, the information contained in this book is sold without warranty, either express or implied. Neither the author, nor Packt Publishing, and its dealers and distributors will be held liable for any damages caused or alleged to be caused directly or indirectly by this book.
Packt Publishing has endeavored to provide trademark information about all of the companies and products mentioned in this book by the appropriate use of capitals. However, Packt Publishing cannot guarantee the accuracy of this information.
First published: July 2017
Production reference: 1280717
Published by Packt Publishing Ltd.
Livery Place
35 Livery Street
Birmingham B3 2PB, UK.
ISBN 978-1-78728-703-7
www.packtpub.com
# Credits
**Author**
Jordan Hudgens
**Acquisition Editor**
Ben Renow-Clarke
**Content Development Editor**
Radhika Atitkar
**Technical Editor**
Joel D'souza
**Proofreader**
Safis Editing
**Indexer**
Pratik Shirodkar
**Production Coordinator**
Arvindkumar Gupta
**Cover Work**
Arvindkumar Gupta
# About the Author
**Jordan Hudgens** is the CTO and founder of DevCamp, where he leads instruction and curriculum development for all the DevCamp and Bottega code schools around the US.
As a developer over the past decade, Jordan has traveled the world building applications and training individuals on a wide variety of topics, including Ruby development, big data analysis, and software engineering.
Jordan focuses on project-driven education as opposed to theory-based development. This style of teaching is conducive to learning how to build real-world products that adhere to industry best practices.
Additionally, Jordan has published multiple books on programming and computer science along with developing training curriculum for Learn.co, DevCamp, and AppDev on the topics, namely Ruby on Rails, Java, AngularJS, NoSQL, API development, and algorithms.
# www.PacktPub.com
# eBooks, discount offers, and more
Did you know that Packt offers eBook versions of every book published, with PDF and ePub files available? You can upgrade to the eBook version at www.PacktPub.com and as a print book customer, you are entitled to a discount on the eBook copy. Get in touch with us at `<customercare@packtpub.com>` for more details.
At www.PacktPub.com, you can also read a collection of free technical articles, sign up for a range of free newsletters and receive exclusive discounts and offers on Packt books and eBooks.
<https://www.packtpub.com/mapt>
Get the most in-demand software skills with Mapt. Mapt gives you full access to all Packt books and video courses, as well as industry-leading tools to help you plan your personal development and advance your career.
## Why subscribe?
* Fully searchable across every book published by Packt
* Copy and paste, print, and bookmark content
* On demand and accessible via a web browser
# Customer Feedback
Thanks for purchasing this Packt book. At Packt, quality is at the heart of our editorial process. To help us improve, please leave us an honest review on this book's Amazon page at <https://www.amazon.com/dp/1787287033>.
If you'd like to join our team of regular reviewers, you can e-mail us at `<customerreviews@packtpub.com>`. We award our regular reviewers with free eBooks and videos in exchange for their valuable feedback. Help us be relentless in improving our products!
> To my sweet and loving daughter, Kristine. I cherished every moment of writing this book at coffee shops all over the country with you!
# Preface
_Skill Up: A Software Developer's Guide to Life and Career_ is an all-purpose toolkit for your programming career. It has been built by Jordan Hudgens over a lifetime of coding and teaching coding. It helps you identify the key questions and stumbling blocks that programmers encounter, and gives you the answers to them! It is a comprehensive guide containing more than 50 insights and methodologies that you can use to improve the work you produce, and to give advice in your day-to-day career.
Focusing on your life skills and the key soft skills we need in the modern world, _Skill Up: A Software Developer's Guide to Life and Career_ will help you find your path to being a better and a happier coder.
# What this book covers
Part 1, _Coder Skills_ , contains advice for people starting out in a coding career, or those who are already working as in a programming role but want to improve their general skills. It includes such subjects as how to study and understand complex topics, defining deep work and what it means for developers, and getting past skill plateaus when learning new languages.
Part 2, _Freelancer Skills_ , contains advice for developers working as freelancers and trying to manage their careers and bid on new tenders. It includes such subjects as knowing when to fire a client, practical tips for taking over legacy applications, and a guide to automating client update messages.
Part 3, _Career Skills_ , contains advice for having a successful career as a developer. It provides information on how to advance your career, and practical tips, such as interview guides. It includes such subjects as how to practice programming techniques and improve as a developer, how to balance best practice and creativity as a developer, and developer salary negotiation strategies.
# Who this book is for
This book is useful for programmers of any ability or discipline. It has advice for those thinking about beginning a career in programming, those already working as a fully employed programmer, and for those working as freelance developers.
# Conventions
In this book, you will find a number of text styles that distinguish between different kinds of information. Here are some examples of these styles and an explanation of their meaning.
Code words in text, database table names, folder names, filenames, file extensions, pathnames, dummy URLs, user input, and Twitter handles are shown as follows: "Returning to our case study of memorizing CSS elements, let's look at the border attributes available in CSS3: `border`."
**New terms** and **important words** are shown in bold. Words that you see on the screen, for example, in menus or dialog boxes, appear in the text like this: "Here in the image I would move a task from being a **To-Do** , to being **Assigned** , to **Working** , to **Under Review** , and finally to **Finished.** "
# Reader feedback
Feedback from our readers is always welcome. Let us know what you think about this book—what you liked or disliked. Reader feedback is important for us as it helps us develop titles that you will really get the most out of.
To send us general feedback, simply e-mail `<feedback@packtpub.com>`, and mention the book's title in the subject of your message.
If there is a topic that you have expertise in and you are interested in either writing or contributing to a book, see our author guide at www.packtpub.com/authors.
# Customer support
Now that you are the proud owner of a Packt book, we have a number of things to help you to get the most from your purchase.
## Errata
Although we have taken every care to ensure the accuracy of our content, mistakes do happen. If you find a mistake in one of our books—maybe a mistake in the text or the code—we would be grateful if you could report this to us. By doing so, you can save other readers from frustration and help us improve subsequent versions of this book. If you find any errata, please report them by visiting <http://www.packtpub.com/submit-errata>, selecting your book, clicking on the **Errata Submission Form** link, and entering the details of your errata. Once your errata are verified, your submission will be accepted and the errata will be uploaded to our website or added to any list of existing errata under the Errata section of that title.
To view the previously submitted errata, go to <https://www.packtpub.com/books/content/support> and enter the name of the book in the search field. The required information will appear under the **Errata** section.
## Piracy
Piracy of copyrighted material on the Internet is an ongoing problem across all media. At Packt, we take the protection of our copyright and licenses very seriously. If you come across any illegal copies of our works in any form on the Internet, please provide us with the location address or website name immediately so that we can pursue a remedy.
Please contact us at `<copyright@packtpub.com>` with a link to the suspected pirated material.
We appreciate your help in protecting our authors and our ability to bring you valuable content.
## Questions
If you have a problem with any aspect of this book, you can contact us at `<questions@packtpub.com>`, and we will do our best to address the problem.
# Part I. Coder Skills
# Chapter 1. Discovering the Tipping Point for Developers
If you've been programming for a while, a question that has most likely crossed your mind is this:
> _"Am I a good developer?"_
Before we go on, let me share a little secret with you... Every developer, even senior developers, have insecurities when it comes to programming. Few individuals like to share that information, mainly because confidence and even arrogance has become a developer stereotype for some stupid reason.
However, I won't BS you. I can tell you that the more experience I have as a coder, the more I realize how much more there is to learn and how far I still have to go.
# Tipping point for developers
With all that being said, I want to discuss topic of defining the tipping point for developers, which is essentially the point at which a developer goes from a beginner to a pro. Since this topic is a bit abstract, it's not possible to point to a specific point in time and say:
> "Here it is, this is when it all clicks and makes sense."
There's not a sentinel moment when programming mastery occurs. It's different for every individual.
## My own experience
I remember when I was originally learning programming. Understanding the syntax and context did not come easy for me. It seemed like I spent 99% of my time looking things up and copying and pasting code from others just to get my programs running.
## The doubt machine
Needless to say, my confidence as a programmer was very low in the beginning. I kept being plagued by nagging doubts, such as:
* Maybe programming isn't for you
* Even if you code works you won't be able to write your own programs
* You're only typing in what the book is saying to do, you won't be able to build anything custom
And the negative thoughts continued from there
## The painful process
If you're a new developer maybe some of this sounds familiar to you, or maybe it doesn't and I simply lacked confidence. Either way, I trudged along, trying everything I could think of to improve as a developer:
* Going through dozens upon dozens of programming books in various languages
* Trying to build up a portfolio of project
* Following online guides
However, back when I was originally learning how to code, the online resources weren't quite as good as they are today.
## The tipping point(s)
So, what did the trick and pushed me over the edge to become a professional developer? None of those things... and all those things. I persevered through project after project and I consumed every training resource I could find. And slowly something amazing started to happen:
> _"Everything started to make sense."_
### The first tipping point
Even though it was a while ago, I still remember the moment my first development tipping point happened. I was sitting in front of my computer in a coffee shop and working on a web application.
A few hours went by and I stopped dead in my tracks, realizing that I had just spent the afternoon building a project and hadn't looked up a single code snippet. It wasn't like I programmed the space station, the project was incredibly basic. However, it was one of the most exciting moments I can remember in my life.
### The second tipping point
As great as that was, I still had very far to go. I remember the next moment when I felt like I reached another key milestone. Even though my confidence had increased as a developer, the thought of anyone seeing my code was a very scary thought. However, I had started to build my freelance business and a client (who was also a developer) asked me to perform a pair programming session with him.
He had run into a bug with the program we were building and asked me to jump on a screen sharing session where we could work on the project at the same time. Honestly, I was scared to death when he asked. I had never coded in front of anyone before and the thought of doing it with this client pretty much gave me a panic attack. However, I didn't really have a choice in the matter so I started the session with him. After a few minutes of nervousness, I started to relax and to my surprise not only did I not make a fool of myself, I actually figured out the bug in his code and got the feature working.
### The secret
So, what was my secret to getting over the hump and going from a beginner to a professional developer? Unfortunately, there is no easy-to-follow recipe. However, there is a process that is guaranteed to work. And the process isn't specific to becoming a programmer, it's the same whether you want to be a developer or a professional athlete... it's hard and smart work.
### The book
In the book _The Tipping Point_ , by _Malcolm Gladwell_ , Gladwell gives countless case studies of what it takes for individuals to achieve mastery in a specific field. The key comes down to how dedicated an individual is to a specific skill. The book postulates that it takes around 10,000 hours for an individual to become a true master of whatever they're pursuing.
I'm not sure I believe in the 10,000-hour rule, mainly because there are a large number of variables when it comes to learning a topic or skill and rarely does a single rule apply for all fields. Also, I think the quality of your practice makes a significant difference.
For example, if you're learning how to play the violin: 5,000 hours of practice with a world class instructor is probably equivalent to 10,000 hours trying to figure it out yourself. However, with all that being said, one thing cannot be denied, _the key to mastery is hard work_.
# The solution
I'm sorry if you were hoping for a quick fix. I can tell you from experience that there are no shortcuts to becoming a developer. You need to learn:
* The fundamentals of coding
* How to build projects on your own
* Various process for working through bugs
Becoming a great developer is not an easy road. However, be comforted in the fact that you are 100% in control of how skilled you will become. The formula is straightforward: the harder you work, the better you will get. So, get your hands on all the material you can find on the language and framework you want to learn. Work through challenging applications and you will be well on your way to mastery.
And soon you will be able to have the exciting moment of clarity when everything starts to click.
# Chapter 2. Are Developers Born or Made? – Debunking the Myth of Prodigies
When talking to development students, I've discovered one topic that constantly arises in conversation. And that topic is the misconceived notion that great developers are born with a special programming gene. So, let's walk through the question _are developers born or made_ , from a practical perspective.
# Are prodigies real?
Before tackling this question, let's take a step back and discuss the topic of prodigies. Because whenever someone thinks that a certain group of individuals are born with superhuman-like talent, they're essentially saying that these special people are prodigies.
## The Mozart case study
But are prodigies real? Let's take a look at one of the most famous prodigies of all time, Mozart. At the age of 5, Mozart was playing concert grade music to the royal family. Surely, this would qualify Mozart as a prodigy, right?
In his book, _Peak: Secrets from the New Science of Expertise_ , researcher _Anders Ericsson_ dispels a number of commonly held prodigy myths. He had this to say about Mozart:
> "If you compare the kind of music pieces that Mozart can play at various ages to today's Suzuki-trained children, he is not exceptional. If anything, he's relatively average."
In his book, Ericsson dedicates a full chapter to debunking the concept of prodigies. And in each case, he illustrates that the individuals achieved their respective levels of success through massage amounts of work.
## Are developers born or made?
Extending the Mozart case study, let's discuss how this applies to developers. Whenever we see a skilled coder it's easy to think that they were born with the innate ability to build applications and that learning new languages and frameworks comes easy to them.
However, nothing could be further from the truth. Over the years I've known more developers than I can count and I have yet to find a single one that was a born developer. I know programmers that work for Google and Amazon, along with computer science professors who specialize in research that boggles my mind to think about. And as amazing as all of these individuals are, each one of them became a great developer through hard work and dedication.
## The tipping point
In Chapter 1, _Discovering the Tipping Point for Developers_ I've discussed the tipping point for developers. The longer I teach and the more I work on my own coding skills, the more I'm convinced that the key to excellence is as straightforward as focused practice.
If you want to become a skilled developer badly enough, and you're willing to:
* Dedicate the time
* Learn from experienced teachers
* Fight through frustrating challenges
* Continually build projects with features you've never developed before
You're going to wake up one day and realize that everything is clicking and that you've become a professional programmer.
# Why we love the prodigy myth
Before I end this chapter, I want to address a subtle issue that explains the reason of why we, as humans, love the idea of prodigies.
The concept of prodigies, individuals born with a natural ability to be successful at a certain skill, such as sports, math, or programming, can be detrimental to our own success. This belief is dangerous because it causes our minds to have negative responses to failure.
For example, if you're an aspiring developer who thinks that programmers are born and not made, when you come across a bug that you can't seem to figure out or a feature you can't build, your first reaction might be:
_I guess I wasn't born to do be a developer._
Or:
_I wish I had talent like XYZ programmer, everything seems to come so easy to him._
If you catch yourself with thoughts like these, remind yourself that prodigies aren't real.
Developers achieve greatness through smart work and perseverance. The 10,000-hour rule from the _Tipping Point_ book by _Malcolm Gladwell_ may not be exactly accurate. However, it does give a general guide for how much work is required to reach a level of mastery.
If you feel like you weren't born with the "developer gene", go and knock out 10,000 hours of focused practice on building applications. I think you'll be pleasantly surprised to find that you'll become so good, that other people will look at you, and they'll think... that you were just born this way.
# Chapter 3. Do You Have to Be a Genius to Be a Developer?
We've discussed the topic of whether great developers are born or made. And in this chapter, we're going to look at a similar topic from a different angle. And we're going to answer the question _do you have to be a genius to be a developer?_
Because of the near-magical nature of coding, one of the most common remarks I hear from individuals who hear what I do is:
_"Oh wow, you're so smart!"_
In fact, just recently I traveled to meet with a group of developers and the head of the company introduced me by saying:
_"This is Jordan, he's just here to be smart."_
I know that when people say things like this it comes from a good place. However, it bothers me. And it bothers me for a couple reasons:
1. First and foremost, these type of comments make it seem like all it takes to become a great developer is being smart.
2. Following up on the above point, these remarks devalue the countless hours of work that are required to learn development.
# The running man
I'm going to get off my soap box for a moment and discuss the life of Steve Prefontaine. If you've never heard of him before, Prefontaine was one of the world's greatest runners during his time. Before tragically dying in a car accident at 24 years old, he had already broken seven track world records.
During his climactic rise to success, many people would try to compliment Prefontaine by saying how talented he was, and by calling him a prodigy, such as on the cover of Sports Illustrated.
However, he was famous for getting furious at people for this type of statement. He said that his success had literally nothing to do with talent. In fact, he said he wasn't born with any innate ability as a runner. Instead he credits 100% of his success with his legendary work ethic.
# Do you have to be a genius to be a developer?
It's important to take the same approach that Prefontaine took as developers. If you fall into the trap of thinking that only geniuses can become good coders, it will most likely lead to quitting when tasks become challenging. This is because our minds constantly are searching for ways to work less. And if you believe that being a genius is a requirement for development, you will have a built-in excuse for faltering on your developer learning journey.
## The way the mind works
In a comprehensive educational study published in Scientific American (<http://www.scientificamerican.com/article/the-secret-to-raising-smart-kids1/>), kids were broken into two groups and taken through some academic assignments. Both groups scored around the same for the assignments. One of the groups were praised by their parents and teachers, and the focus of the compliments centered around how smart and talented the kids were.
The second group of students were complimented in a different manner. Instead of complimenting students on their innate ability, students were complimented on how hard they worked.
After going through this cycle of compliments, the same two groups of students were presented with new, and very challenging assignments.
The first group of students, the ones that had been told that they were brilliant, ended up giving up and not completing the tasks that were assigned to them. However, the second group of students, the ones that were complimented on their hard work, performed dramatically better than group 1.
### The reason
So why did two groups of students have such different results when, by all appearances, the students had the same level of actual skill?
The researchers concluded that the students from group 1 felt like the top priority was maintaining their _genius_ status. So, they quit the second assignment early because they didn't want to look bad and tarnish the _genius_ label that they had been given.
However, the second group of students didn't feel the pressure to maintain a genius status. Instead, they wanted to maintain their new reputations as hard workers. With this mindset, the second set of students worked through the challenging topics instead of giving up.
### A smarter approach
So, instead of taking the mindset that you need to be a genius to become a developer, take the mindset that best developers are also the hardest working developers. With this approach, your potential for skill is literally limitless. You are 100% in control of how good you will become as a coder. And your success will completely be determined how hard (and how smart) you are willing to work.
# Chapter 4. How to Study and Understand Complex Topics?
When I was younger I used to struggle learning a new or difficult subject, and over the years and about a decade of university and grad school have helped me put together a strategy for how to study and understand complex topics. Typically, I apply this learning system to subjects such as algorithms and software engineering; however, it can be applied to any topic.
While there is a near infinite set of study strategies out there, I like this approach because it utilizes a divide and conquer strategy, focusing on breaking a complex topic into easy-to-understand components, and putting the pieces back together at the end to see how they all work together.
# A system for how to study
Let's take a case study example: understanding how logarithms work. Logarithms are used throughout the fields of mathematics and computer science; however, unless you use them regularly it's easy to get rusty on them:
1. The first task that I will do is take a piece of paper and write Logarithm in the center and circle it.
2. Next, I'll go to a comprehensive post on the topic, such as one on Wikipedia. When reading the first sentence, I come across a few terms that are a bit fuzzy:
* Inverse operation
* Exponentiation
I will stop reading the logarithm article and go and read those two articles until I feel comfortable with what they represent. After I feel good about those two items, I write them as their own circles that connect to the Logarithm circle. I will also add any examples that will help me understand what the terms mean if necessary.
3. Next, I'll go back to the original Logarithm post and keep going through the article repeating this process until the entire page is filled with a mind map that explains each component that makes up logarithms and how they work together. This may include base case examples, such as:
_64 = 2^6 is the same as log 2 (64) = 6_
If this seems like a dead simple approach to study...it is. The goal of studying is to learn a topic, and one of the easiest ways to understand a complex subject is to break it into easy to comprehend components. For example, if you're trying to understand an advanced algorithm in computer science from scratch, you may feel a little intimidated.
However, if you break the algorithm down into small enough components you'll see that it's essentially a process of steps made up of connecting simple modules such as loops, manipulating variables, and using conditionals. A problem is only hard when you try to think of it as a whole. However, any concept can be understood if you simplify it down to easy to comprehend pieces.
Obviously, the more complex the topic, the longer it will take to deconstruct; however, I am a firm believer that anyone can understand any topic assuming they dedicate themselves and put the work in. I hope that you can leverage this mind mapping process to understand complex topics and that it will help you learn how to study properly and truly learn.
# Chapter 5. Effective Study Practices for Developers
Let's imagine that you're back in school and midterm exams are coming up. How would you study? Some common approaches might be:
* Re-read the study materials or lecture notes
* Highlight and memorize the key terms
* Go over your notes constantly until test day comes
Those all sound like effective study practices. However, cognitive research has shown that many of the traditional study patterns that students have followed for decades simply do not work.
I didn't make up that list of study patterns. That's exactly what I used to do in preparing for exams. However, I discovered (after failing a number of tests) that these strategies failed miserably when it came to helping me to truly learn new concepts.
# Why traditional study habits don't work
This type of approach to studying doesn't work because our minds don't function like computers. A computer can take in information and then spit it back out. However, our minds are more relational in nature.
By relational in nature, I mean that our brain functions like a graph-based network. If new information attempts to enter the brain without being connected to any of our previous knowledge, it will simply be rejected.
For example, let's imagine that you are new to learning programming. If you simply run through a list of programming terms and syntax rules, you might memorize them in the short run.
However, because your brain hasn't been properly introduced to the concepts, it will eventually eject the information, viewing it as useless since it's not related to the rest of your view of the world.
However, imagine that you take a different approach. In this new, more enlightened approach, you work with your brain and allow it to connect each of the new programming concepts that you're learning to knowledge and experiences that you already have.
# An effective study practices case study
Whenever I'm teaching a new programming concept to students, I try to draw a fitting analogy to a real-world concept. This process is called reification and I view it as one of my most important tasks as a teacher.
Let's imagine that you are learning about the MVC (Model, View, Controller) design pattern in software development. You could take the approach of trying to memorize each of the roles of the Model, View, and Controller. However, that strategy wouldn't help you answer questions related to how each of the components work together. And if you memorize quiz questions and answers, you probably will have issues answering anything that you haven't memorized.
## The reification example
What if instead of trying to memorize key terms about the MVC pattern you focused on drawing a real-world analogy to the process? My favorite way to understand this type of architecture is comparing it to a restaurant:
* **Model** : The model is the chef in the kitchen. In the same way that a chef prepares the meal for customers, the model works directly with the data for the application.
* **Controller** : The controller works like a restaurant waiter. In an application, the controller's role is based on taking requests and managing communication between the model and the view. This is much like a waiter who takes customer orders, communicates with the chef, and eventually brings the food out to the table.
* **View** : The view is like the table that a customer is sitting at. It doesn't do much besides provide a platform for placing the food on. This is exactly like how the view should operate in an application. If built properly, a view should simply be a place where data is shown to users.
Do you see what we just did? We learned about the MVC design pattern in a way that our minds can actually comprehend. I could fall out of bed and recite back the role of each component of the MVC architecture, not because I spent countless hours trying to memorize them, but because I connected the concept to my real-world experiences.
# The hard way
Over the years I've concluded that if studying is easy...I'm doing it wrong. I used to follow study pattern of:
1. Read
2. Memorize
3. Repeat
This was partly because it was easy. It wasn't mentally taxing to sit down and read through a textbook or my notes. However, research is proving that this type of study habit is not only ineffective, it is also damaging.
## Additional negative effects
How is it damaging? If you have followed this type of study system you know one thing: it takes time. This time spent reading and memorizing could have been used in countless other ways that would have proven more effective in the long run. And when it comes to studying, time is one of the most valuable assets that you have, so wasting it is akin to an educational felony.
# The comprehensive study system
In addition to the process of reification, there are a number of other study strategies that research is showing to be more effective than traditional study practices. In their book _Make It Stick_ , cognitive psychologists _Brown, Roediger_ , and _McDaniel_ give the following recommendations for studying:
* When learning from a textbook, use the key terms from the back of each chapter to test yourself.
* List out key terms and use each one in a paragraph; this will test to see if you understand a concept outside of the realm of how the textbook or instructor supplied it.
* While reading new material, convert the main concepts into a series of questions and then go back and answer the questions when you're done reading the chapter.
* Rephrase the main ideas in your own words as you go through the material.
* Relate the main concepts to your own experiences, much like the reification process we've already discussed.
* Look for examples of key concepts outside of the text. When I'm learning a new programming language I never rely on a single source. If I come across a concept that doesn't make sense I'll usually review 2- 3 other sources that provide alternative explanations to what I'm attempting to learn.
# Summary
In summary, when it comes to effective study practices, make sure that you're making the most of your time. Remember that the most important goal with studying is retaining knowledge so that you can use it in real-world scenarios. And the best way to accomplish this goal is by following strategies that work with your mind's learning patterns.
# Chapter 6. Defining Deep Work and What It Means for Developers
Standing on the podium, Michael Phelps stares at the American flag and listens to the National Anthem after winning gold once again. After watching Phelps win 21 gold medals (at the time I'm writing this), it's natural to ask: "Was he simply born for greatness?" I don't know. Yes, his body type has helped him take advantage of physical elements of swimming.
However, there are millions of individuals with his height and wingspan who watch him at the Olympics from their couches every four years. There is no magical swimming gene that Phelps was born with. Instead, the secret to his success can be found in his discipline to a practice called _deep work_. Muscle Prodigy (<https://www.muscleprodigy.com/michael-phelps-workout-and-diet/>) research claims:
> "Phelps swims minimum 80,000 meters a week, which is nearly 50 miles. He practices twice a day, sometimes more if he's training at altitude. Phelps trains for around five to six hours a day at six days a week."
If Malcom Gladwell's 10,000-hour rule is even close to being accurate, Michael Phelps surpassed this benchmark years ago.
In case you're wondering how this applies to coding, don't worry, I haven't forgotten that this is a show for developers.
# Definition of deep work
As you go through these chapters, you may discover that one of my favorite books is _Deep Work_ by _Cal Newport_. (The fact I referenced the book a few dozen times may given it away). So, what exactly is deep work? A dead simple explanation of deep work is:
> _"Deep work is the ability to focus without distraction on a cognitively demanding task."_
Whether you believe that swimming is cognitively demanding or not, I believe that Phelps's example is fitting. If you have ever attempted to train with the level of intensity that Phelps does, you can attest to the mental toll that training takes. So essentially, deep work can be simplified by saying that it has the following characteristics:
1. It's a real-world action. It's not a theoretical concept, it's something that you can practically implement.
2. To work properly you have to eliminate 100% of your distractions.
3. The task has to be challenging.
## The deep work strategy for developers
Let's dissect the definition of deep work and build a practical strategy for how it can be implemented from a developer perspective. Let's imagine that you want to learn about the computer science topic of asymptotic analysis. If you've never heard of asymptotic analysis, don't worry, you can trust me that it qualifies as a challenging topic.
### Taking action
Let's start with the fact that deep work is an action. With that in mind, you will need to make a clearly defined time slot. If you have never practice deep work studying before, I'd recommend limiting the slot to around two hours. As you'll discover deep work is a draining task. For our example, let's imagine that you have designated 9 AM to 11 AM as when you're going to study asymptotic analysis.
### Removing distractions
With your time slot set, now it's time to remove any and all potential distractions. Let me be 100% explicit; this means:
1. You cannot check your email.
2. No phone calls. In fact, put your phone in airplane mode to ensure no one calls or text messages you.
3. Don't even think about checking Instagram, Facebook, Twitter, or Tinder. All your swipe rights will have to wait for a few hours.
If I missed any distractions you can add them to the list. It may also help to listen to classical music to block out any potential sound distractions while you study.
### Study hard and smart
Now that you have dedicated a few hours to studying asymptotic analysis and have removed all your distractions, it's finally time to get down to business. If you think that now you can simply start reading a few Wikipedia posts, I'm sorry, that won't earn you a deep work badge.
For deep work to be truly effective, it has to be difficult. If I was learning about asymptotic analysis for the first time and wanted to practice deep work while studying it, I'd take the following approach:
1. I'd begin by reading a number of online resources on the subject.
2. Next I'd watch an online lecture while taking notes.
3. I would then find practice exercises where I would attempt to figure out problems from scratch.
4. Next, I would write a blog post or record myself teaching the concept.
5. Lastly, I would have another student or instructor review my teaching and exercises to ensure that I understood the concept properly.
Do you see how much more comprehensive this type of studying is? Even if you had never heard of asymptotic analysis before your deep work study session, you would be fluent in it after you were done.
# Multiple sessions
When I mentioned earlier how you should limit your deep work sessions to around 2 hours, I don't mean that you can understand any topic in that period of time. Some complex topics may take days, weeks, months, or years to properly understand. So, it is completely fine to spend a number of sessions working through the same concept. If you are going to do this, I recommend that you make notes for what you were doing when you stopped. This will allow you to pick up right where you left off.
# Summary
I hope that this has been a helpful introduction to what deep work is and how you can practically implement it as a developer. If you want to learn more about the topic I suggest that you pick up Newport's book. It will give you a great set of tools for learning how to use deep work to constantly improve as a developer. When it comes to learning, deep work is the closest thing you can get to steroids. Good luck with the coding!
# Chapter 7. Task Switching Costs for Developers
In this chapter, I'm going to discuss the concept of **task switching costs**. Task switching, commonly referred to as **multitasking** , can be detrimental to your performance as a developer and can even lead to errors in your projects. Our world has changed dramatically over the past decade, whether for good or bad is not a topic we'll discuss in this chapter. However, one thing is sure: we are constantly bombarded with distractions.
As I was researching this chapter, I received over a dozen emails, 7 Snapchat messages, 30 notifications on Instagram, 7 Twitter notifications, 5 Skype instant messages, and surprisingly only 9 text messages. If you were counting, that's around 72 various notifications that were pushed to me in the past two hours. Beyond that, I researched this chapter at a coffee shop filled with potential distractions.
So exactly how bad are distractions? Research from Gloria Mark (<https://www.fastcompany.com/944128/worker-interrupted-cost-task-switching>), who is a Professor in the Department of Informatics at the UC Irvine, shows that it takes, on average, 23 minutes and 15 seconds to get fully back on task after being distracted. That's a very, very bad thing when it comes to productivity; however, I've seen it myself, I've lost track of how many times I'll be in the middle of a development project and receive an email on a completely unrelated matter and instead of ignoring it and continuing to work I'll read it and then spend time working on another task before returning to the project.
This may not sound like a major issue, except that when I come back to the project, I don't pick up from where I left off. Instead I have to re-familiarize myself with what I was working on the moment that I was distracted. If the problem was complex, it may take me even longer than the 23 minutes in order to get back in the zone and working on the project.
So, in a world filled with emails and social media distractions, how can anyone get any real work done? After reading Cal Newport's book _Deep Work_ , I started to put together some practical ways that I can work efficiently and still stay in touch with the world.
# A system for decreasing task switching costs
1. If I'm working on a project, I set aside a specific amount of time that morning. For example, if I'm working on Project X for 2 hours, I will put it on my calendar and say that from 9 AM to 11 AM I'm working on Project X.
2. I remove any and all negative distractions during that time. That means I'll usually put my phone on Airplane mode so I don't receive any social media notifications. Notice how I said negative distractions? I made this distinction because in the same research report from UC Irvine it revealed that not all distractions are bad. If the distraction is related to the task that you're working on, it can actually be beneficial. For example, if I'm working on the routing engine for a web application and the client messages me to discuss the application, what they say may actually influence the work that I'm doing or give me an idea on how to refine it. That's a good distraction and it's why I typically will keep my email and instant messenger on while I'm working. However, if I see that the Skype message or email is coming from another client or is completely unrelated I'll simply ignore it. I do know many Deep Work proponents who would say that 100% of your distractions have to be eliminated; however, that's not always practical.
3. Have a clear conclusion for whatever you are studying or working on. If you don't establish an end for the task, your mind is going to be prone to wander in the same way that a runner without a finish line won't be able to effectively compete in a race. The research around task switching costs also reveals that even planned distractions are harmful, so if you are planning on working for 2 hours straight on a project, don't plan any breaks in the middle of the task. Maintain your focus throughout the allotted time and then you'll be free to relax afterwards.
I hope that this has been a helpful overview of task switching costs and that you now have some practical methods for staying on task.
# Chapter 8. How to Use Willpower Limits Instead of Letting Them Use You?
There are a number of common characteristics among great developers. However, few virtues are as important as willpower. World class coders constantly are forced to work through complex concepts, and without willpower they would give up before experiencing any kind of breakthrough. In this chapter, I'm going to walk through the topic of willpower limits.
This will include a practical walk through on:
* What willpower limits are
* How you can improve your personal willpower limits
* A plan for being intentional about managing willpower limits
# What are willpower limits?
For graduate school I have performed extensive research on the topic of task switching costs. While studying about task switching, I came across the topic of willpower limits and how they related to performance. Essentially, the study of willpower limits says that _individuals have a limited amount of decision making power each day_.
## How many decisions do you make each day?
If that sounds weird to you, don't worry, I had a hard time with the concept right away too. So, let's go through a typical day for a developer. What are some decisions that you make each day?
* Deciding to get up or press snooze on the alarm clock
* Picking out what to eat for breakfast
* Selecting your clothes for the day
* Deciding if you want to go to the gym, a run, or walk around and play Pokémon Go
* Deciding on which route to take to work
* And the list goes on and on
Notice how none of those items are related to development at all. And in fact, those were all common decision items that you have to make each morning before you even get into work. If you actually count the number of decisions that you have to make each day, you'd discover the number is probably in the hundreds or even thousands. If you include subconscious decisions such as staying in your lane while driving, the number is most likely in the millions every day!
## Why is willpower important?
Hopefully, I've helped you see all of the decisions that we make daily. So why do willpower limits matter when it comes to making decisions? Mainly because without willpower the quality of our decisions will suffer dramatically.
Imagine yourself without willpower for a second. With no willpower, you:
* Would eat whatever you wanted, harming your overall health
* Wouldn't study, thus never improving as a developer
* Would recklessly spend money on whatever popped into your mind, forcing you into debt and eventually bankruptcy
It's not a pretty picture, which is why willpower is so important when it comes to making decisions. Willpower gives you the self-control to make the right decision, even when it's not the easy one.
## Are willpower limits real?
So, with all of that in mind, is there really a limit to the amount of willpower you have each day? I recently went through the book, _The Willpower Instinct_ , written by _Dr. Kelly McGonigal_ (no relation to Professor McGonagall that I'm aware of). In the book Kelly presents research and countless case studies that clearly show that we do indeed have a limit to our daily willpower.
Imagine that your willpower is like a bottle of water. Each morning you start with the bottle filled to the top. And each time you make a decision or have to use willpower, such as deciding to get up instead of hitting snooze on the alarm clock, a little of the water gets poured out. As you go through your day you'll eventually "pour out" all of your stored-up willpower.
## When the willpower well runs dry
So, what happens when the willpower well runs dry? Typically, it leads to poor behavior, such as:
* Procrastination
* Making bad decisions
* Poor performance
If you find yourself experiencing these types of thought patterns, it could very well be due to hitting your willpower limit too quickly.
I know from experience that I typically write my best code in the morning when I'm fresh, whereas I find myself running into more development bugs when I work later in the evening. When I realized this pattern, it made me believe even deeper in the concept of willpower limits and how they can negatively affect performance.
## Saving up willpower
With all of this in mind, the concept of saving up our willpower reserves seems like a pretty important concept. Let's go back to the water bottle analogy. If you were in a desert and had a limited supply of water, what would you do? I think the obvious answer is that you would only use the water when it was needed.
So, if we treat our willpower like a precious resource, it would make the most sense to use it on our most important tasks each day.
What's a practical way of doing this? Let's walk through a simple but practical example.
### One outfit to rule them all
If you watch my show on CronDose you may have noticed something... You get a gold star if you noticed that for the last 13 weeks (14 weeks if you include this week) I've worn the same shirt. Please note, it's not the same exact shirt. When I decided to experiment with the _one outfit_ concept I purchased eight identical shirts.
So why am I doing this? By wearing the identical outfit each day, it completely removes the set of decisions that I would normally have to make each morning when I'm getting dressed. I no longer have to pick between 100+ shirt and jeans combinations. And it has the added benefit that it's quite comfortable.
Does wearing the same outfit each day really help improve my performance? I can't scientifically say one way or the other. Most likely it has a negligible effect. However, it has a much more powerful benefit than simply removing my morning outfit decision. Each day when I put this shirt on it reminds me that I have a limit to my willpower and that I need to use it wisely. And having that mindset does make a difference.
### Being a copycat
As a side note, the idea of wearing the same outfit is not an original idea. Steve Jobs, President Obama, and Mark Zuckerberg all have a similar ritual and that's where I got the idea from. If some of the most successful individuals in the world make it a priority to remove any and all unnecessary decisions, I thought it would be a good idea to try out.
## Focusing willpower
Wearing the same outfit each day is a good idea for taking care of some low-hanging decision-making fruit, but it's only the beginning. To really ensure that you get the most out of your willpower each day, you need to be intentional with how you use it.
For example, I've talked before about how I have a daily list of to-do items that I follow religiously. To ensure that I'm getting the most out of my day, I always schedule my most challenging tasks in the morning. By taking this approach, I don't risk running out of willpower while I'm working on a vital project. From there I save my lower priority to-dos, such as reading, for the end of the day.
By following this pattern, I've noticed a significant improvement in my work over the past few months and I also feel more relaxed at the end of the day.
# Summary
I hope that this has been a helpful discussion on the topic of willpower limits and that it has given you some ideas on how to manage your own willpower.
# Chapter 9. Cramming Versus Consistent Study and a Study System that Works
In this chapter, I'm going to discuss the concept of cramming versus consistent study. And don't change the channel if you're not in school, if you're a developer, or if you want to learn software development the learning never ends.
On an average, I typically go through over a dozen books at the same time and around 4-5 various online courses because the deeper I get into development, the more I realize how much more I really need to understand.
With that in mind, I think the topic of cramming versus consistent study habits should be beneficial since the way that we study is just as important as the volume of how much we study. Most of us have been in the situation where we put off studying for too long and before we know it an exam is upon us that we have to cram for. If you can remember back to the last time that you crammed for an exam or project, how much of what you studied can you remember today?
If you're like me, probably not much. While I was in college I was very bad at this and ended up cramming for many of my midterms and finals, with mixed results from a grade perspective. However, once I got to computer science graduate school at Texas Tech I ran into a problem—cramming didn't work at all.
Software development concepts build upon themselves, so what was taught in the Fall semester would be the foundation for even more complex topics that would be discussed in the Spring. In the Fall, I would learn about logic programming and in the Spring, I'd have a course where I had to build a production application using the Prolog programming language.
Using cramming as a study technique resulted in me having very poor retention of what I was learning, which meant I had to go back and relearn the topics that I had already forgotten from the previous semester. I don't have to tell you how stressful this made my academic life, not to mention the fact that I was working as a full-time developer at the same time. So, I knew that something had to change and I put together a system for helping me retain what I learned each day through a consistent study pattern. Much like a function in programming, my system for consistent study takes in a few parameters:
* Scheduling
* Fighting procrastination
For scheduling I created a to-do list, segmented by day, for what I needed to study, which included academic papers, books, and watching online lectures. I put these in a drag and drop to-do list on Basecamp. After I studied a particular item I would drag it up to the next day's to-do list so I would have a visual cue that I was done for that day.
For me, I would procrastinate studying because staring at the list of the books I had to read was intimidating, and this was mainly due to the fact that I didn't set any practical goals for studying. If you stare at a Discrete Mathematics textbook and tell yourself to study, it's natural to want to put it off; however, if you set small goals, you're less likely to put it off.
With that in mind, I'll put a note, such as read 3 pages of my Information Retrieval textbook, and 3 pages doesn't sound nearly as scary as the vague "just study" mindset. The interesting result in making small, manageable goals for studying is that not only does it help curb procrastination, but typically I will also read much more than the 3 pages. There have been plenty of times where I set of goal of a few pages of a book and ended up reading a few chapters.
With all this being said, there are times where I plan deep work study sessions. In one of these sessions I will set aside 2-3 hours of time to sit down, without distractions, and work through a complex topic. However, I always limit the time to no more than 2-3 hours per day, and I will usually not study any other topics on these days since I'm usually mentally drained by the end of them.
I hope that this chapter has been helpful and will help you develop your own system for studying so that you can retain when you learn and be able to use it when it matters most.
# Chapter 10. Is Reading Important for Developers?
Throughout this book, I have written quite a bit about improving as a developer, specifically discussing various ways to study from a practical perspective. However, in this chapter, I want to specifically answer the question: _is reading important for developers?_
The short answer to the question is: yes! However, as computer scientists it's poor form to simply take someone at their word. So, let's dive into why reading is critical to improvement.
# Why is reading important for developers?
Let's analyze a few key statistics with regard to reading.
## CEOs and reading
How many books do you currently read a year? If your answer is that you're too busy to read entire books, let me ask you another question: are you busier than the CEOs of the world's most successful companies? Probably not.
However, research from the Rype Academy (<http://rypeapp.com/blog/5-easy-ways-to-read-more-books-and-double-your-knowledge/>) shows that CEOs such as Elon Musk, Mark Cuban, and Peter Thiel read around 60 books a year! That's 4-5 books each month.
## Compounded learning
So why do some of the most successful individuals in the world take the time to go through so many books? At a high level it may seem excessive, but if you truly believe that knowledge is power, wouldn't it make sense to dedicate whatever time is needed to attain more knowledge?
If you look at reading like a form of linear learning, then yes, reading would be a waste of time. Linear learning would be a 1 to 1 transfer of knowledge. For example, if it took the author of the book 10 years to research a topic and it took me 10 years to go through the book, that would be pretty pointless. At the end of the day this type of reading would be pointless.
However, I look at reading like it's compounded learning. What is compounded learning? Good question! Compounded learning is the process of taking the knowledge from an individual, but not having to spend the same amount of time that it took that individual to research the topic.
### A compounded learning case study
For example, imagine that you read a book on _How to Become a Better Developer_. The author of the book had to spend years researching the topic (assuming that it was a well-written/well-researched book). However, if you go through the book in a few weeks, that means that you were able to gain years worth of knowledge in a few weeks!
Research (<http://blogs.plos.org/neurotribes/2011/06/02/practical-tips-on-writing-a-book-from-22-brilliant-authors/>) shows that top authors will spend a minimum of two years researching a book. And that research time doesn't take into account the fact that authors draw on their entire lifespans to write a book. All of this means that each time you read a book it's as if you were able to gain a lifetime's worth of experiences and wisdom from the author.
## The CEO who didn't have time to read
A few years back, I was offered a CTO position for a startup in New York City. The job had a good salary, great stock options, and an excellent product.
However, during a dinner meeting with their Founder/CEO, I asked him about a book I had finished reading that discussed best practices for tech startups. He said that he had never heard of the book.
This wasn't a problem, there are millions of books and I don't judge someone for having different literary tastes than myself. However, the CEO followed this statement up by saying that he didn't have time for reading. He was too busy building the business.
The CEO's view of reading resonated with me during the job consideration process. And I ended up turning down the job. If the CEO didn't dedicate time to read and learn from others, that means that he would be relying solely on his own knowledge and life experiences. And even the most brilliant business person will fail if they think that they already have all the right answers.
## My reading system
It's one thing to say that reading is important; it's another thing entirely to go through a large number of books on a regular basis. With that in mind I've developed my own reading system. This system also takes into account a number of complaints that I've heard others say about reading.
## The reading schedule
First and foremost, I schedule a set amount of time each day for reading. Usually, this equals around 1-2 hours; however, on weekends this number can be double that number. At any given point of time, I'm usually going through a dozen books ranging from mind/skill hacking through technical programming books.
## Audio books are books too!
I'm not sure where the stigma of audio books came from. However, with my travel schedule, I've discovered that audio books are an invaluable tool in my learning arsenal. Obviously, you can't go through programming books via Audible. However, you can go through skill and business-based books. And I personally have hundreds of books in my Audible account, many of which I've gone through multiple times. In fact, many of the books I've discussed and quoted from were books I listened to rather than read.
## Books are too expensive
One of the top complaints I hear from students is that books are too expensive. My response is always: if you're not willing to sacrifice to improve, then you're not going to attain your goals. And that includes sacrificing financially.
With that being said, there are ways that you can go through a large number of books, even if you're on a budget. To start off, your local library has countless books that you can learn from each day. And assuming that you bring the books back on time, a library is a completely free option. I have a library within walking distance of my home in Scottsdale, AZ, and I will visit it a few times a week to discover new books.
Additionally, you can sign up for book memberships. Safari Books Online offers an _All You Can Read_ package. I have this membership and have gone through a large number of technical programming books in their database over the years.
# Summary
In summary, is reading important for developers? I believe that it is. Reading enables you to activate compounded learning. And if you have the chance to gain years' worth of knowledge and experiences in a few weeks, it seems insane to pass up on an opportunity like that.
# Chapter 11. Learning How to Code – Getting Past Skill Plateaus
A common pattern I see with students learning how to code is:
1. Quickly learning a massive amount of information.
2. Then running into a seemingly insurmountable wall. In this phase, the student typically feels like they've reached the zenith of what they're going to be able to accomplish about development.
This second phase is called a plateau. In this chapter, we're going to walk through strategies for getting past skill plateaus.
It's important to understand that everyone follows a similar pattern when it comes to learning a new skill. This means that you will experience times where it seems like every day you're soaking in a wealth of new information. But it also means that you will run into times where it feels like your mind will limit you from learning anything new.
# What is a learning plateau?
When it comes to hitting a learning plateau, it's important to look at the potential root causes for why it's occurring. It's been my experience that no two plateaus are the same. And until you've diagnosed why you're not learning, you won't be able to move on to your next level of skill.
## False ceiling
Before I continue I want to reiterate something: _you will never reach a point where your level of skill is maxed out_. Maybe if you're a professional athlete and getting older, then your body is naturally going to decrease in performance. But when it comes to concepts such as understanding development, if you continue to dedicate yourself and if you're willing to listen to experts, your skill will never reach a peak.
# Getting past skill plateaus
Over the years I have witnessed a few key reasons why individuals (and myself) run into skill plateaus.
## Proper information/resources
When a student lacks access to proper information, it makes learning a more arduous process. Imagine a talented developer in high school who had been relying on her teacher (who had limited skill). In cases like this, the student will need to find additional resources, such as online courses, that will help teach her concepts she's never been taught before.
## Best practices
During a phase of the learning cycle when best practices are the focus, students may feel like they are hitting a learning plateau. I remember when I was first learning about test-driven development. The concept seemed counterintuitive. I would spend 2-3 times the amount of time on a feature. And this became incredibly frustrating. It felt like I wasn't learning anything new because my new knowledge wasn't affecting anything on the screen.
However, this phase isn't actually a skill plateau. There are many times where developers need to take a step back and focus on quality over quantity when it comes to building applications.
My advice for going through this phase is to embrace it. Be patient. As soon as you have a firm understanding of how the best practices can be utilized, you'll be able to move onto learning new concepts. The only difference is that now you will be able to leverage your new skills, the result being that you'll be a more refined developer.
## Challenging/new tasks
In my experience, the main cause of students hitting a skill plateau is when they stop challenging themselves. If you remember back to when you were first learning development, it seemed like your knowledge was skyrocketing each day.
The reason for this was because each of the concepts you were learning were completely new to you. However, after a certain period of time it seems like it's natural for us to want to become comfortable. Instead of trying to learn something new each day, we simply try to duplicate the work that we've done up to a certain point.
This approach is less taxing mentally. However, it has the nasty side effect of limiting how we improve. Whenever I feel like I'm getting into a rut, I will look at popular websites and I'll start to put together a list of features that I want to learn how to build. From that point, I can put a plan together for what concepts I need to learn in order to implement them.
## Frustration = skill
One of my favorite illustrations of getting past skill plateaus was made by the calligrapher, Jamin Brown:
Notice in this illustration how the learning process is filled with plateaus? This is a natural component when it comes to improving at any skill.
But also notice that the key to overcoming a plateau is called the **Frustration Zone**. I think that's a great name for it. Learning complex topics is not easy. As you've probably heard countless times, "if it were easy, everyone would do it".
Becoming a developer can be one of the most rewarding experiences that someone can have. And part of what makes learning how to code so fulfilling is how many challenges you'll need to overcome to succeed.
# Summary
I hope that this has been a helpful guide and that you now have some practical strategies for getting past skill plateaus. And good luck with the coding.
# Chapter 12. Developer Learning Curve – Why Learning How to Code Takes So Long
When it comes to becoming a developer, one of the questions I get asked the most is: why does it take so long to learn how to code? I've discovered the answer can be found in research related to learning curves.
# What is the learning curve?
The concept of learning curves has been around since 1885. Typically, the research has been performed in the psychological and cognitive sciences. However, the concept can be clearly utilized when it comes to learning development.
# The developer learning curve
The following graph shows the standard learning curve. This was generated by a big data analysis algorithm that analyzed the learning patterns of individuals in a number of industries. The curve is smooth because it takes the average learning process and averages the process.
Later in this chapter, we'll take a look at what a learning curve looks like for a single person. Over the years I've had the privilege of teaching students how to become developers. I've witnessed this learning curve play out again and again. And in this chapter, I want to examine the three stages that all developers go through. Additionally, I'll discuss about how long it takes to traverse from one stage to another. The three stages that I'll discuss are:
* Liftoff
* The twilight zone
* The zone
# Liftoff
Let's start off by taking a look at the liftoff stage. This is an exciting time for new students. During this time students are immersed in learning skills that they've never seen before:
Because all the topics that students learn during this stage are new, their expertise skyrockets. I like to call this the liftoff stage because it's easy to visualize a new student's expertise like a rocket ship soaring into the sky into places it has never been before. During this time, a student will learn how to:
* Configure a development machine
* Learn a programming language
* Work with various frameworks
* Build functional applications
This stage usually lasts for the first 250-300 hours that a developer is learning how to code. This estimate is based on what I've seen with the DevCamp bootcamp students and equals around 6-8 weeks of intensive learning.
As fun as this stage is, it has drawbacks. One of the key problems is that it can give students false confidence. When they see themselves building applications that actually work, it's natural to believe that they can dive right into building production apps for clients. However, they don't realize that they're about to enter... the twilight zone of learning how to code.
# The twilight zone
After the exciting liftoff stage of the developer learning curve, aspiring developers will enter the twilight zone:
This is a challenging time for students and many students decide to quit programming entirely during this stage.
So why is this time so challenging? After seeing countless students go through it, I've discovered that there are a number of contributing factors:
* While in this stage, many of the core concepts and commands haven't cemented themselves in a student's long-term memory. This results in them having to constantly look up documentation, query Stack Overflow, and things like that.
* During this time, the novelty of simply having an application work has worn off. Now students are asked to perform advanced tasks such as:
* Working with legacy applications
* Debugging defects
* Improving performance
* Building features that they don't have a step-by-step tutorial for
* Additionally, while working through the twilight zone, students are expected to start implementing best practices. In the launch stage, the primary goal was to get applications functional.
During this next phase, students start learning how to build applications that can be used in real-world scenarios. This means that a student may spend five times longer to build an application with the identical feature of something they created during the launch stage.
This can be frustrating; however, the increased time spent implementing best practices allow the applications to be scalable and flexible enough to be used in production. This is in stark contrast to the apps created during the launch phase that don't adhere to industry standards.
# The zone
There is good news though; if a student persists through the twilight zone of learning they will enter The Zone of the developer learning curve:
This zone is entered usually after around 1,000 hours of study and work. During this stage, developers have a wide range of features they can build without having to look up the documentation.
In this stage, when you visit Stack Overflow you'll be answering as many questions as you ask. And thankfully, learning new concepts will come easier. The reason why learning is easier at this stage is because you will have developed a mental model of development.
For example, I recently started working with the Scala programming language. I've been able to pick up on how to build applications in Scala dramatically faster than when I started learning C or PHP a decade ago. This is because I have a decade of knowledge in the development space that allows me to frame the new concepts. When I read the documentation and see what it says about data types, I don't have to wonder what a data type is. Instead I can skip ahead to learning the syntax.
As you'll notice in the developer learning curve, the growth pattern in this phase is less than the other two stages. As you've heard me say countless times, learning never ends for developers. However, learning does change. During this phase, a developer focuses on learning topics such as:
* Incremental performance improvements
* Building helper code libraries
* Refining how application code flows
# A unique journey
Throughout this chapter you may have noticed that the developer learning curve was smooth. However, that's not reality. The reason why the curve was smooth was because it averaged out the learning path of a large number of individuals. When it comes to a single student, the learning curve looks more like the following graph:
There are ups and downs throughout the learning cycle. As a student, you may decide to switch programming languages after a few years (like I did when I switched from PHP to Ruby around 5 years ago).
Even though you don't have to start back from scratch, it will still take time to learn a new language or framework. And throughout your development journey you'll discover plenty of ups and downs. So, don't get discouraged if you aren't satisfied with your skill level, because I have a secret to tell you: good developers never feel like they've arrived and are done learning.
# Summary
I hope that this has been a helpful guide to understanding the developer learning curve, and good luck with the coding.
# Chapter 13. Slowing Down to Learn How to Code Faster
Nowadays, it seems like everyone wants to do things faster. We want to pay without taking out a credit card or cash. Social media lets us share images and videos from our lives in a split second. And we get frustrated if Netflix takes more than 3 seconds to start streaming our latest TV show series binge. However, if you want to learn how to code faster, I'm going to present an odd idea: **go slower**.
This may seem like a counterintuitive concept. After all, don't coding bootcamps, even DevCamp where I teach, tell you how you can learn how to code in a few months? Well yes, and research shows that 8 weeks is a powerful number when it comes to learning. The Navy Seal training program specifically chose 8 weeks as its timeframe for conditioning candidates. And if you search the web for the phrase 8 Week Training programs, you'll find courses ranging from running 10ks to speaking Spanish fluently.
So yes, I'm huge believer that individuals can learn an incredible amount of information in a short period of time. But what I'm talking about here is becoming more deliberate when it comes to learning new information.
# Learn how to code faster
If you're like me, when you learn a new topic the first thing you'll do is either move onto the next topic or repeat the concept as quickly as humanly possible. For example, when I learn a new Ruby or Scala programming method I'll usually jump right into using it in as many different situations as possible. However, I've discovered that this may not be the best approach because it's very short-sighted.
## Our default mind
When it comes to learning how to code, one of the most challenging requirements is moving knowledge from our short-term memory to our long-term memory.
Remember the last time you learned a programming technique. Do you remember how easy it felt when you repeated what the instructor taught? The syntax seemed straightforward and it probably seemed like there was no way you would forget how to implement the feature. But after a few days, if you try to rebuild the component, is it easy or hard?
If you're like me, the concept that seemed incredibly easy only a few days ago now causes you to draw a blank. But don't worry. This doesn't mean that we're incompetent. Instead, it means that this piece of knowledge wasn't given the chance to move from our short-term to our long-term memory.
## Hacking the mind
So, if our default mindset is to forget what we've learned after a few days (or a few minutes), how can we learn anything? This is where our brain's default programming comes into play and where we can hack the way that we learn.
I'm currently teaching myself the TypeScript programming language. TypeScript is the language that is recommended for Angular 2 development, so I thought it would be a good next language to learn. However, instead of taking my default approach, which is to slam through training guides and tutorials, I'm taking a more methodical approach.
### Slowing it down
Through my learning path, I'm going through a number of books and video series. And as I follow along with the guides, as soon as I learn a new topic I completely stop. I'll stand up. Write the new component on one of my whiteboards. And actually, write the program out by hand.
After that, I type the program out on the keyboard... very slowly. So slowly that I know I could go around 4-5x faster. But by taking this approach I'm forcing my mind to think about the new concept instead of rushing through it. When it comes to working with our long-term memory, this approach is more effective than simply flying through a concept because it forces our minds to think through each keystroke.
### Bend it like Beethoven
I didn't learn this technique from another developer. Instead, I heard about how one of the most successful classical music institutions in the world, the Meadowmount School of Music in New York, taught students new music compositions. As a game, the school gives out portions of the sheet music. So, where most schools will give each student the full song, Meadowmount splits the music up into pieces.
From there, it hands each student a single piece for them to focus on. From that point, the student will only learn to place that single piece of music. They will start out very slowly. They won't rush through notes because they don't even know how they fit into the song. This approach teaches them how to concentrate on learning a new song one note at a time.
From that point, the students trade note cards and then focus on learning another piece of the song. They continue with trading cards until each student has been able to work through the entire set of cards.
By forcing the students to break a song into pieces they no longer will have any weak points in a song. Instead, the students will have focused on the notes themselves. From this point, it's trivial for all the students in the class to combine their knowledge and learn how to play the song all the way through.
### From classical music to coding
So, can this approach help you learn how to code faster? I think so. The research shows that by slowing down and breaking concepts into small pieces, it's easier for students to transfer information from the short-term to long-term memory.
# A practical system
So, the next time you are learning a coding concept, take a step back. Instead of simply copying what the instructor is teaching, write it down on a piece of paper. Walk through exactly what is happening in a program.
If you take this approach, you will discover that you're not longer simply following a teacher's set of steps, but that you'll actually learn how the concepts work. And if you get to the stage of understanding, you will be ready to transfer that knowledge to your long-term memory and remember it for good.
# Chapter 14. Mental Models for Learning How to Code and Improve as a Developer
I've talked quite a bit about what it takes to become a great developer. To achieve a level of mastery, I've discussed a number of criteria and in this chapter, I want to add a new pre-requisite to the list.
Let me begin by asking you a question. If I showed you some code, would you be able to tell me in a few seconds if it's good or not? The world of software development is incredibly complex. However, I've discovered over the years that the best developers have the uncanny ability to instantly judge the quality of someone's code.
I spoke to you in Chapter 2, _Are Developers Born or Made? – Debunking the Myth of Prodigies_ about the notion that prodigies and savants are a myth. But if this is the case, how can expert developers analyze programs so quickly? To answer this question, we need to go back to Fake Ancient Greece.
# Mental models for the Kouros
I said Fake Ancient Greece because my favorite illustration of mental models was discovered alongside one of the greatest forgeries in modern art history.
In Malcolm Gladwell's book _Blink_ , he tells the story of the Greek Kouros. In 1985, the Getty Museum purchased a Greek statue called the Kouros for over $9 million dollars. Initially, the museum was hesitant to purchase the statue because there was a significant fear that sculpture was a fake. Kouros pieces were so incredibly rare, the chances that a legitimate and well cared for piece had been discovered were slim to none.
However, the museum was willing to take the risk and embarked on a fact-finding mission. They put the statue through every scientific test available at the time. And the Kouros passed with flying colors. After going through the full examination, the museum purchased the Kouros for $9 million dollars.
Art historians from all over the world were flown in for the unveiling of the Kouros. But something went terribly wrong. The moment that these specialists saw the statue they knew the Kouros was a fake. Interestingly enough they couldn't give any actual reason.
**They simply knew that something was not quite right**. Their suspicions turned out to be correct and the Kouros ended up being proved to be a hoax. But how were these individuals able to do what countless scientific studies could not? It all comes down to mental models.
## What are mental models?
In preparation for this chapter, I was discussing the topic of mental models with a friend and was surprised when she looked at me, confused. After informing me that she'd never heard of mental models, I decided to add this section to explain what mental models are. And after that we'll get into how we can build them to learn development.
A mental model is a mental representation of a specific topic or skill. You can't create a mental model overnight or with cram sessions. Mental models are developed through years of repetition and countless hours of honing a craft.
My Dad is a major league hitting coach for the Houston Astros. Throughout my life I've been able to watch him instruct hitters on how they can improve their swings.
And I'll never stop being amazed by the fact that he can watch a new hitter's swing and within a split second pick out multiple ways that the player can improve. I can tell you that he did not develop this skill in a short period of time. He has spent more time watching hitters and film than anyone I know. And over the years he has developed a mental model of what the perfect swing looks like.
## Mental models for developers
OK, so we've talked about art historians and baseball coaches, but how can we create mental models as developers? You may or may not like the answer, but it doesn't really matter because it's the truth. Mental models are made through repetition.
However, repetition by itself isn't enough. For example, if you built an identical program every day for 10 years you would get really, really good at building that one application. However, you wouldn't improve as a developer. I've talked before how medical research shows that doctors who have spent years practicing on the same types of patient are less proficient than doctors fresh out of residency. In the same way, as developers we improve when we're stretching ourselves each day.
You can stretch yourself by doing things such as:
* Learning a new programming language or framework
* Teaching others how to learn programming
* Creating an open source code library and allowing other developers to use it
Einstein said it best when he said:
> "The only source of knowledge is experience."
# Summary
If you dedicate enough time each day improving yourself as a developer, you will be able to truthfully answer yes to the question I posed at the start of this chapter. You will be able to have the ability to look at a piece of code and instantly know if it's good or bad. And you'll know that it's not some type of coding super power; instead, it's a skill that you earned through your constant pursuit of improving as a developer.
# Chapter 15. A Developer's Guide for Hacking Procrastination to Achieve Success
There you are. Sitting in front of your computer. Staring at a blank screen. You know you have to work on a code project, but it feels like you're frozen. The task before you is so intimidating that you don't even know where you begin. It feels as if you'd rather be doing anything else in the world besides that task that's staring you in the face.
This scenario is the ugly and all-too-common face of procrastination that programmers are forced to fight constantly. If this situation sounds familiar, you're in good company. But if you want to become a professional developer, you'll need to implement a system for **hacking procrastination**. And that's what we're going to walk through in this chapter.
As the lead instructor for DevCamp I get asked questions from students from around the world. However, one of the most prevalent inquiries I get from aspiring coders is how to overcome procrastination.
# Root causes of procrastination
Before we walk through a system for hacking procrastination, we first need to dive into the root causes for this negative habit. Everyone is unique, but over the years I've seen procrastination is typically caused by three thought patterns:
* Perfectionism
* Fear of success
* Lack of planning
To overcome procrastination and get back on track we'll need to address each one of these issues. Because if you let any of these mindsets control the way your mind operates, you will never be able to reach your potential.
# Hacking procrastination
I called this chapter hacking procrastination because I think that **hacking** is the most appropriate term for what needs to happen to achieve success. Developers hack applications to build features or fix bugs. In the same way, we need to hack our thought patterns so that our brains function properly.
Before we go through the system I want to make one concept clear. As humans, we were made for action. Procrastination is a negative habit that we've learned through fear-driven thought patterns. To be successful at anything in life, whether it's development or business, overcoming procrastination is a requirement.
## Hacking perfectionism
Starting off the list of the causes for procrastination is perfectionism. Have you ever watched a baby trying to stand up for the first time? Babies, who haven't learned that failure is a bad thing, will spend countless hours trying to stand up.
Each time they fall down it doesn't seem to faze them in the slightest. But you won't find a baby that lets perfectionism get in the way of achieving their goal. Instead, they will keep trying until they can stand up and eventually walk by themselves.
However, somewhere between the time that we're babies and adults we develop the thought pattern that we're not supposed to fail. So instead of trying and failing until we succeed, we simply try to only perform tasks that we know we can do properly. To hack perfectionism, we have to remove the component in our brain that is afraid of failing.
If you are a developer learning how to build a new feature that you've never worked through before? Let me clear something up. You are going to do it wrong the first time!
And that's 100% fine. If you think that by waiting you are magically going to learn how to perform the task perfectly, you are sadly mistaken. So, step one is: embrace failure and remove the requirement of perfectionism.
## Hacking the fear of success
Next on the list is hacking the fear of success. If you're overcome the trap of perfectionism, congratulations. However, I've seen just as many developers get stuck due to the fear of success as the fear of failure.
This concept may seem odd since success doesn't seem like something that you should be scared of. However, I remember when I was first learning development. When I was walking through a coding book I would get so excited when I discovered a new concept. However, then I would freeze. My mind's first response was:
> "If you learn this, then what are you going to do?"
For example, when I first learned how to build a connection to a database, I put the book down and didn't pick it up until weeks later. By learning the database concept, it opened up a new and scary new world of all of the new topics I had to learn after that. All of a sudden, I had to understand:
* SQL queries
* How to build relationships between database tables
* SQL injection requirements
* And the list seemingly went on infinitely in my mind
To hack the fear of success, we need to quieten our minds. The fear of success is really rooted in the fear of the unknown. So, whenever you're feeling this fear, simply take a step back. Be happy that you have learned a new topic. And then move onto the next feature or topic.
Don't let your mind run wild with all of the potential, unknown concepts that you'll need to learn in the future. Like learning anything else, you need to take it one step at a time.
## Hacking the plan
Last on the list for hacking procrastination is creating a practical plan. When I recognize that I'm procrastinating I now tell myself to look at my plan of attack. Usually I'll discover that my plan is too general.
For example, if I'm building a payroll application, I may have an item on my to-do list that says: **Build reporting engine**. That's a scary feature! That's the type of item that will stick on my to-do list for weeks without me taking any action.
To fix this, I've learned that if I break the requirement into a series of very small tasks I can break the cycle of procrastination. For the reporting engine feature I can create a series of much smaller, more manageable tasks, such as:
* Create a page for users to access reports
* Implement a database query for pulling the reports from the database
* Build a file downloader for reports
When I break a large and scary feature down into small pieces, I instantly feel better. The feature is no longer scary and I no longer feel like putting it off until tomorrow. Instead, I am able to simply follow a set of small tasks each day until the feature is complete.
# Summary
I hope that this has been a helpful guide for helping you break the cycle of procrastination in your own projects and that you will be able to use it to become a more effective developer. I'll leave you with a quote from the book _The Five Elements of Effective Thinking_ by _Edward B. Burger_ and _Michael Starbird_ :
> "Being willing to fail is a liberating attribute of transformative thinking."
So, put yourself out there, create a practical plan, and stop procrastinating and start coding!
# Chapter 16. The Problem with Procrastination for Developers
Libraries could be filled to overflowing with books filled on procrastination. Through my life and career, I have gone through self-help books that attempt to explain why people procrastinate along with supplying strategies to help curb procrastination.
And as great as all those books are, no one has been able to describe the true problem with procrastination better in my mind than _Steven Pressfield_ in his book _The War of Art_.
# The problem with procrastination
In _The War of Art_ , Pressfield compares procrastination with being an alcoholic. If you're like me, when I first heard this comparison I was skeptical. I had a hard time connecting myself pushing off writing a blog post until tomorrow with an alcoholic passed out on the sidewalk in front of a bar.
However, I chose to continue reading. Pressfield gave procrastination a name, calling it the resistance. And that was something I could relate to. Whenever I come across a challenging task, it's as if there is a constant voice in my head saying:
> "Wouldn't this feel great to push to tomorrow?
>
> You'll be excited to do it tomorrow."
And when I give into the voice, it's as if I took a shot of happy pills. I instantly feels as through a weight has been lifted off my shoulders and I feel happy. However, when tomorrow rolls around I've discovered something... the voice comes right back and it's still encouraging me to push the task off again.
## Instant gratification
After going through this cycle of procrastination for years I finally did recognize the pattern. And Pressfield was right, procrastinating on tasks has the same root cause as being an alcoholic. Alcoholics are willing to trade long-term joy for short-term happiness. By this I mean that an alcoholic will risk their health, career, and family, all for the sake of the feeling that a drink will give them at that moment.
This pattern is played out in the mind of all of us when we procrastinate. When we continually put off tasks for tomorrow, we are trading long-term success for short-term convenience.
## Baby steps to knock out procrastination
I've already presented my system for hacking procrastination. However, I don't want to describe a problem without giving a solution. Therefore, I will conclude by saying that the best way I've discovered to fight procrastination is by taking baby steps.
In his book _Mini Habits_ , _Stephen Guise_ made the concept of performing one push up a day famous. Guise was a self-proclaimed procrastinator who despised going to the gym or working out. However, one day he decided he was going to create the mini goal for himself to perform a single push up every single day. By following this approach, he realized that the idea of working out was no longer a scary concept. And therefore wasn't something to procrastinate.
Of course, doing one push up a day would have limited health benefits. But what Guise discovered was that after performing the push up he was usually in the mood for doing more pushups. And eventually, his daily habit morphed into a full daily fitness regime.
### Baby coding steps
I've discovered the same approach works with learning and development. When there is a task that I feel like pushing off, I tell myself that I only have to work on it for thirty minutes.
By giving myself a doable goal, the task is far less intimidating and therefore, I don't feel the same resistance and desire to push it off. And typically, I'll discover that I want to work longer than 30 minutes on the task. The end result being that I get much more done and I no longer fear my daily to-do list.
# Chapter 17. Practical Ways to Use the Pomodoro Technique as a Developer
As we continue to work through ways to hack the developer's mind, the focus of this chapter is going to be on increasing productivity. Specifically, we're going to analyze practical ways to use the Pomodoro Technique.
I'm constantly researching new ways to improve my personal productivity. And through my journey as a developer, a popular approach that I've discovered is the Pomodoro Technique. This is a process that I've utilized and I credit it with allowing me to focus on a large number of tasks each day.
Don't let the weird name scare you away. The Pomodoro Technique is a dead simple productivity system that focuses on splitting tasks into timed intervals throughout the day.
# Practical ways to use the Pomodoro Technique
One of the greatest strengths of the Pomodoro Technique is how easy it is to implement. The process that I follow is:
1. Each morning I pick out the tasks that I want to accomplish that day.
2. I then decide how long each task will take. The Pomodoro Technique works on point system. Each time you work through a 25-minute task you earn a point.
3. Typically, I try to earn 10 Pomodoro points each day. This means that if I have 3 tasks that I know will take an hour each, I will earn 6 points for those tasks. And it means that I have 4 additional 25-minute slots available for the rest of the day.
## Taking a break
Did you notice how I kept saying 25-minute time slots? There is a reason for the odd number. The Pomodoro Technique places a high priority on taking scheduled breaks. After completing each 25-minute task, you take a 5-minute break. During this free time, you can do anything you want. You can get on social media, you can take a walk around the block, or anything that you want to do. Just make sure that your break does not exceed 5 minutes.
Also, after you've completed 4 tasks it's recommended that you take a 15-minute break. However, you can tailor your breaks and intervals to what works best with your schedule. By planning breaks throughout the day, you will decrease your chances of burn out. And I've noticed that I no longer feel bad about doing things such as checking my Instagram account or Hacker News throughout the day, because I can fit my guilty pleasures into my scheduled free time.
This is one of the aspects that I truly love about the Pomodoro Technique. Many of the other productivity systems I've tried in the past tend to lead individuals towards burning out. However, the Pomodoro approach allows you to have a sense of balance.
## Lifestyle versus fads
Have you ever tried dieting before? When I was younger I struggled with my weight and to help fix it, I tried a number of intense diets. This included nutrition strategies such as dramatically decreasing calories, or killing off carbs. However, I noticed that I'd stay true to the diet for a few weeks or even a few months, but eventually I would fall back into poor eating habits.
Once I recognized this trend I moved to having a balanced approach to eating. I stopped trying nutritional fads and I transitioned my focus into eating in a way I felt I could eat for the rest of my life.
I made this change in my nutritional approach a few years ago and it's completely stopped my roller coaster dieting and weight loss and weight gain.
### A lifestyle of productivity
In the same way when I was younger I fell into the same pattern with working on tasks. I'd get excited about working on a project or learning a new programming language. And I would spend countless hours working on what I wanted to accomplish.
However, this approach inevitably led to burning out and large stretches of time where I didn't want to work at all. I look at the Pomodoro Technique in the same way that I look at having a balanced diet. By limiting the number of tasks that I work on each day and by implementing planned breaks between each task, I no longer burn myself out.
Additionally, after I have finished my work for the day and have earned my 10 Pomodoro points, I feel a sense of accomplishment that I never felt before. And after work, I don't feel guilty spending time with my family and friends, because I know that I completed every task that I set out to work on that day.
# Practical implementation
So how can you implement the program? There are a few ways. To start off, you can simply use the timer on your phone and then count up each of the tasks/points that you achieved each day. That's how I started off working with the Pomodoro Technique.
Additionally, there are a number of smartphone apps that have Pomodoro timers and even allow for creating a task list that you can use as a pick list for your tasks each day. I like these types of apps because they also give you historical analytics so you can see how many tasks you've completed each day. The Pomodoro focus app (<https://itunes.apple.com/us/app/pomodoro-time-focus-timer/id973134470?mt=12>) is my personal favorite (and it's free).
# Chapter 18. The Power of Making Mistakes – Learning by Failing
Let's take a step back in time back to my first semester of Computer Science grad school. Stepping into my first class I was filled with nervous excitement. The class was taught by Dr. Gelfond, one of the most respected individuals in the artificial intelligence sector.
As class progressed I witnessed a disturbing trend. Instead of simply lecturing us like our other professors, Dr. Gelfond constantly called students up front to write programs on the chalkboard or to describe a concept he discussed. This wouldn't be a big deal, except that he made a habit of calling us up front specifically when it was clear that we did not understand the concept. Was he cruel? Did he want to make us look ignorant in front of the entire class?
# The secret weapon to mastery – making mistakes
Actually, the opposite was true. Instead, Dr. Gelfond cared enough about us that he imparted to us the secret weapon to mastery: making mistakes. Wait, making mistakes is the opposite of what our mind tells us to do, right? Making mistakes is embarrassing. Mistakes tell the world that we don't understand a concept. However, making mistakes also provides a number of powerful tools that anyone interested in learning should be aware of.
## Making mistakes – memory steroids
First and foremost, when you make mistakes, especially publicly, you're going to feel like you're taking memory steroids. How so? When I think back to Dr. Gelfond's class I still remember every mistake I made when I was called in front of the class. The memories generated by making mistakes are so vivid that they can be recalled, even years later like mine. Now obviously simply remembering the mistakes by themselves would be pointless.
However, in addition to remembering what I did wrong, more importantly I remember what I had to do to correct my mistake. It's been over three years since I took that class, but I can still remember each of the key concepts that he taught us. And I can tell you from experience that I cannot say the same thing about all of the classes I've taken.
## Mistakes force learning
Another benefit to making mistakes is that they force you to learn. No one likes being wrong. So, assuming that you have a passion for knowledge, you can use the memory of making mistakes to help motivate you to learn a concept properly.
If Dr. Gelfond would have simply stood in front of the class and lectured for the entire semester, I most likely would have studied enough to do well on the tests and leave it at that. However, because I constantly had the thought in the back of my mind that I may have to be called up in front of the class to write a program or describe a concept, it forced me to study harder than I would have for a test. This healthy fear took me from simply being able to remember a concept to truly mastering it.
## Mistakes kill pride
Lastly, making mistakes helps to kill pride. Proverbs 16:18 says:
> "Pride goes before destruction, a haughty spirit before a fall."
One of the largest obstacles to learning is pride. Anyone puffed up with pride will find that their learning progress will come to a halt. When someone is filled with pride they can't see beyond their own limited knowledge. Thankfully, if you embrace the process of learning by making mistakes, pride will never be able to stake a claim in you. By their very nature mistakes force you to realize that you don't know everything, and that you have more to learn... which we all do.
# Summary
So, whether you are just learning to code from scratch or if you're a seasoned developer, never be afraid to make mistakes. Mistakes reveal that you're traversing into new territory that you've never been before, which is what you need to do to go from mediocrity to mastery.
# Chapter 19. Learn How to Code – The Guide to Memorization
During a recent bootcamp teaching session where I was walking through a number of frontend development techniques, a student asked a great question. Referencing the CSS styles, she asked:
> "What is the best way to remember all of the specific style names and properties?"
This is a vital question to answer, especially for new students. For example, if you look at the CSS documentation you'll find thousands of potential style options. If you're learning these styles for the first time that list can be pretty intimidating. And that doesn't even bring in the idea of learning how the styles work together with applications as a whole!
Obviously, this issue does not only apply to CSS styles. When it comes to learning development, whether it's a programming language or framework, you will be greeted with a large amount of information that you'll need to memorize, or at least know where to reference it.
# The guide to memorization
At first glance, this may seem like a daunting task. And many aspiring developers have given up on their learning journey because it seems like an insurmountable challenge.
However, I'm here to tell you that it's completely realistic for you to learn how to work with a large number of complex concepts. And if you follow the system I outline in this chapter, you'll be amazed at how quickly you pick up on memorizing more information than you ever thought possible.
## Repetition
Before I go into the memorization system I have used over the years, it's important to say that repetition is the key to memorizing large amounts of information. None of the techniques I will give you are going to help if you don't take the time to work through them consistently.
## Smarter, not harder
With that being said, it's important to know that, by itself, repetition is a slow and naive memory training technique. As a development student, imagine that I had a list of a few hundred method names and tell you to memorize them. If you were to simply stare at the sheet of paper and try to memorize the names, how do you think you'd do? If you're like me and the majority of the world, probably not very well.
The reason why dry repetition isn't a great way to memorize names is because it doesn't give you a frame of reference for the names.
### Visual mental mapping
In the first memory technique, we're going to walk through visual mental mapping. Our minds are incredible at memorization. However, at the same time, our minds are also picky with how they store information. Let's run a quick test. If I show you 15 random digits, such as:
* 234
* 348532
* 984
* 234523
* 34534
* 35234
* 234
* 25345
* 234
* 985
* 553
* 37434
* 740
* 423
* 9812
And I give you 5 seconds to look at each number. How many of the numbers will you repeat back to me? Unless your name is Dustin Hoffman, you probably won't be able to name very many!
However, what if I showed you the pictures of 15 celebrities? Now if I give you the same test as with the numbers, do you think you'd do a better job remembering the list of celebrities or the random numbers? Assuming you know who the celebrities are, you'd be able to repeat back a significantly larger number of celebrities than numbers.
The reason for this difference is because you have a frame of reference for the celebrities and in this exercise, you had a visual reference. By combining these two things your brain was fully prepared to recite back a larger number of items from the second list.
With this knowledge in mind we can apply the same principles for memorizing anything.
### Short-term versus long-term memory
Because our brains are efficient machines they naturally sort information based on priority. You are most likely aware that you have short-term and long-term memory. This concept is the reason why you can instantly remember your second-grade teacher's name decades later, but may forget a new acquaintance's name 30 seconds after hearing it.
Typically, the brain doesn't log knowledge into our long-term memory bank unless it thinks we're going to need it in the future. This is kind of like how a computer works. If you add text to a document and save the file on the hard drive, that's like storing information in the mind's long-term memory.
However, if you run a calculation in the terminal the computer processes the information in memory and then discards it, which is like how our short-term memory system works.
### Implementing visual mental mapping
So, when it comes to implementing the visual mental mapping technique, we're essentially tricking our brain into thinking that it needs to move a piece of information into long-term memory. In this process, we associate a visual image with the term that we want to memorize. A key prerequisite for this to work is that the visualization needs to be relevant to the term (or the behavior of the term).
Getting back to the developer's initial question. Let's see how we can use visual mental mapping to memorize a CSS style. I'm going to use the `text-decoration` property as a case study. In the world of CSS, the `text-decoration` element allows you to add or remove an underline style to a piece of text. With this in mind, I would create an image in my mind that would look something like this:
So, in this example I have an image filled with decorations. And on top of the image, I have some text that is underlined. And it's sitting on the decorated fireplace mantle. By creating this visual image, I've mapped:
* Decoration to underlined text
* A familiar image to something abstract
And with this mental image in place, I don't have to think about the term `text-decoration`, instead I will think of a decorated fireplace with underlined text sitting on the mantle. This visual is much easier for my brain to accept into long term memory because it has a direct frame of reference.
The text-decoration word is no longer a foreign element trying to invade my memory. Instead, it's catching a ride on an image that already has a home in my long-term memory.
### Taking a real-world example
Sticking with our celebrity theme. Imagine that you wanted to go to a private, VIP party in Hollywood. If you just try to show up the bouncer at the door most likely won't let you in. However, if you're friends with Brad Pitt and you walk in together, you won't have any issues attending the party.
Visual mental mapping follows the same principle. Our brains guard our long-term memory to ensure that our mind doesn't get cluttered with useless information. For example, what if you logged every piece of information that you come across each day into your long-term memory?
As you drive down the street to work your brain captures millions of data points, such as street signs and people walking, etc. If your brain didn't guard against useless information entering your long-term memory bank, all of this information would be treated with the same priority as your parent's names. Obviously, this wouldn't be a good idea!
So, our brains are like the guard in the VIP Hollywood party. And when we attach a new piece of information to something already logged in long term memory, it's like we're having Brad Pitt escort us into the party.
### Finding patterns
So visual mental mapping seems like a great idea. However, the idea of creating thousands of visualizations isn't very practical, which is why, when I'm learning a new programming language, I also focus on picking up on patterns.
Returning to our case study of memorizing CSS elements, let's take a look at the border attributes available in CSS3:
* `border`
* `border-bottom`
* `border-bottom-color`
* `border-bottom-style`
* `border-bottom-width`
* `border-color`
* `border-left`
* `border-left-color`
* `border-left-style`
* `border-left-width`
* `border-radius`
* `border-right`
* `border-right-color`
* `border-right-style`
* `border-right-width`
* `border-style`
* `border-top`
* `border-top-color`
* `border-top-style`
* `border-top-width`
* `border-width`
As you can see, there are 21 available attributes. And that's just for managing border styles on a webpage! As you can imagine, it would be pretty intimidating to memorize this list, especially when you realize that it's only a very small percentage of the available CSS styles needed for development.
However, if you start to analyze the list you'll notice a number of trends. For example, there are a number of styles that simply reference: `top`, `bottom`, `left`, and `right`. These styles are simply ways for giving a border style to a specific side of an element.
Additionally, you may also notice that each side also has a set of options for color, style, and width. So practically, if you know that these elements are all available to the border set of elements, this list can be shrunk down to 5 items:
* border
* border-color
* border-radius
* border-style
* border-width
This is more manageable.
### Copy and paste is the enemy
In addition to creating visual mental maps and using patterns, I'm going to finish off the list of memorization techniques with the recommendation to not copy and paste new concepts that you're trying to learn.
I first heard this advice from _Zed A. Shaw_ , the author of the _Learn Hard_ programming book series. He instructs his readers to not even look at the book at the same time that they're implementing the code. He postulates that by forcing yourself to type in the code without referencing the documentation while typing, it forces the mind to actually think through each keystroke.
In my personal experience as a developer and with teaching, I've discovered a significant difference between the students that copied and pasted code or simply followed along with a tutorial, compared with the students that attempted (even unsuccessfully) to implement the code by themselves.
# Not everything has to be memorized
On a final note, I want to dispel a common fallacy. As a developer, you don't have to memorize every class and method to build a project.
Even professional programmers constantly look up documentation on a regular basis. Instead of feeling like you have to memorize everything, focus on memorizing the terms that you use the most. This will make the memorization process more practical and natural.
# Chapter 20. A System for Learning a New Programming Language
In this chapter, I'm going to discuss how to learn a new programming language. I'll walk you through the five steps that I use whenever I'm learning a new language or framework.
Over the years, I've been hired by organizations such as Learn.co and AppDev to write programming curriculums for:
* Ruby on Rails
* Ruby programming
* Python
* Java
* Several JavaScript frameworks
The only language that I really build applications in is Ruby, which means that I've been forced to become proficient in a number of language that I really didn't have much experience working with, sometimes in a very short period of time. And over the years I've developed a system for learning a new language or framework, and that's what I'm going to walk through in this chapter.
When I'm learning a new programming language I follow these steps:
1. Watch a full tutorial series on the language. When I'm watching I don't try to follow along, I simply watch what the instructor does in the demos so I can get a high-level view of the language syntax and flow.
2. Create a _hello world_ application. I'll incorporate a few basics, such as **running a loop** , **creating and instantiating a class** , and any other high-level concepts I remember from the tutorial.
3. Pick out a sorting algorithm and implement it in the language. It's fine if the sorting algorithm is a basic one such as selection or bubble sort. Sorting algorithms force you to use data structures, loops, variables, and functions. Combining each of these elements will give you a good handle on how the language works.
4. Go through an advanced tutorial on the language and this time follow along and build the programs with the instructor.
5. Go through coding interview questions for the language. Being able to confidently answer these questions will give you a good idea if you have a solid understanding of the language.
I've used these five steps for a number of languages and I can also tell you, once you've become proficient in a single language you'll find it's much easier to pick up new programming languages since most of them have quite a bit of shared processes, and all you'll need to do is learn the difference in syntax.
I hope these tips will help you learn a new programming language. Please feel free to write to me with any other methods that you've found helpful when learning, and good luck with the coding!
# Chapter 21. Development Study Tips – Reverse Note-Taking
In this chapter, we're going to go back in time and walk through when I developed the system of reverse note-taking. A quick Google search will show that I have coined the term; however, I did not invent the process.
Back when I started computer science grad school at Texas Tech I was struggling with one of my classes. It had been about a decade since I had been in a classroom environment and I was having a difficult time paying attention to the 1.5-hour lectures.
# The problem with traditional note-taking
During this time, I spent quite a bit of time meeting with Dr. Richard Watson. And during one of our meetings I brought up the issues I was having. His first question was based around how I was taking notes for the course.
I showed him my notes and he instantly told me that I was taking notes completely wrong. He pointed out multiple places in my notes where I had missed key concepts that were unifying elements. And without noting these items, I wouldn't understand the topics at all.
In reviewing the notes, I realized he was completely right. I spent my time writing down facts and what I thought were key terms. However, I regularly failed to articulate how everything worked together.
For example, for my notes on tree data structures I outlined each of the key elements of binary search trees and B-Trees. But I failed to describe the innate differences between the tree components from a behavior perspective.
This is similar to taking notes in a history class and writing down the names, dates, and locations for Napoleon's loss at the battle of Waterloo without describing the critical differences between his old armies with the one he lost with.
## Reverse note-taking
Finding out that I was taking notes wrong was great. But it wouldn't have been too useful without learning an alternative approach. So, Dr. Watson asked me to try a different type of note-taking technique.
He said to put my pen and paper away during class. And instead of taking notes during class, he recommended that I simply listen to the lecture. He instructed that as soon as the lecture was over I should find a quiet place and _then_ write down all the topics that I remembered from the discussion.
Initially, I was skeptical of this approach, mainly because I was afraid that the important concepts would go in one ear and out the other. He added that I should tape record the lecture so that I could use the recording as a safety net for the topics that I failed to remember.
Despite my negative perspective on the approach, I decided to give it a try. (Obviously, my natural note-taking approach wasn't effective, so I didn't have much to lose). I followed Dr. Watson's advice to the letter. And I was pleasantly surprised to discover that I remembered much more information using this reverse note-taking approach compared with simply trying to write down concepts during the lecture.
## Benefits of reverse note-taking
I started following this reverse note-taking process years ago and I still use it today. Through this time, I've noticed a number of key benefits to this approach.
### Narrowed focus
First and foremost, by having the knowledge that I will have to recite back the key components of the lecture forces me to have an increased level of focus. This is opposite to how I used to take notes. My old way of taking notes would many times distract me from the concepts being discussed. I would hear a concept that I felt was important and I would take my focus away from the speaker and focus on writing down the topic.
Many times, this would inadvertently result steal my focus away from another important concept, or a description of how the topic I was writing down worked at a high level.
Additionally, as a naturally competitive person I would make a game of how much I could remember from each lecture. If I could remember enough to write down two pages of notes on Monday, I would try to write down two and a half pages on Tuesday. By making a game of the practice it forced me to narrow my focus even more on the content.
### Story-based mindset
Another benefit of reverse note-taking is that it forced me to think of the lecture as a unified story instead of a series of facts. Let's go back to our illustration of Napoleon's battle at Waterloo. If you listen to a lecture about the battle and take notes during the class, you'd probably do things like write down the following:
* General names
* Cities where battles took place
* Dates
However, if you simply listen intently to the lecture and recite it back afterwards you won't repeat dates and locations. Instead you will naturally remember the battle in story form. You'll discuss the struggles that the Duke of Wellington had to overcome to lead the charge against the French army. And because it's a story, your retention of the topics will be considerably higher compared with attempting to memorize facts and figures.
If I were to ask you to remember a high school history class and a movie you saw in high school, would you have a better chance of remembering the plot of the movie or the history lecture?
So, getting back to my computer science grad school experience, by leveraging the reverse note-taking strategy I forced myself to think of the topics discussed during the lecture as a story as opposed to a bunch of theories and math equations.
### Forced repetition
Lastly, the reverse note-taking approach made it easier to review the lecture material compared with my old style of note-taking. Before that I would rarely listen to a lecture recording. Even if I had the intention to listen to the recording, other priorities always seemed to override the task. I mainly attribute this failure to the fact that I, for some reason, trusted my notes.
However, when I started reverse note-taking I would always listen to the lecture a second time to fill in any items that I missed during my post note-writing session. I discovered this single benefit to be critical to my success since it became an automatic habit to reinforce my knowledge. In contrast to my old approach where I trusted my untrustworthy notes, with reverse note-taking I didn't trust my memory, so I knew I had to reinforce my memory. And the consequence was that I always would listen to a lecture twice, with the final result being a dramatic increase in retention.
# Summary
This approach is not for everyone. I know students who excel with a more traditional note-taking strategy. However, if you find yourself in a situation like mine, I highly recommend you giving reverse note-taking a chance. You may be surprised how effective it can be.
# Part II. Freelancer Skills
# Chapter 22. Tips for Organically Growing a Freelance Business
I strongly recommend that you think about freelance techniques whether you're a freelance developer today or you are working with freelancers in some capacity. You'll gain insights into thinking and coding, whatever type of developer you are today.
Your career will be smarter if you're able to think—at least when it's a help—like a successful freelance developer. That's what this part of this guide is all about: thinking smartly and for yourself as a developer!
We'll begin by discussing how to organically grow a freelance business as a developer. When I say organically, I mean that these are strategies that should work while you sleep. For example, I have had some of my largest clients contact me out of the blue based on them coming across a blog post or GitHub project I published.
# Organically growing a freelance business
Having an organic marketing strategy is key for building a sustainable business. And in my experience the marketing mechanisms that have been the most effective are the six I will discuss.
## Referral requests
Starting off the list are referral requests. Word of mouth marketing is one of the most powerful tools you can use for acquiring new clients. Imagine a real-world scenario of referrals. If someone opens up a new restaurant in your town, are you more likely to be influenced by the restaurant advertisements or one of your close friends telling you how great the food was?
If you're like me, if a friend sings the praises of the restaurant I'm going to, I value their opinion much more than an ad from the restaurant itself. The reason for this is because I trust my friend more than the restaurant. It works the same way when it comes to marketing yourself as a freelancer. If you have happy clients they can help grow your business.
Over the years I've had multiple clients refer their friends and colleagues to me. However, I've also discovered that sometimes it helps to give them a little push. After you have successfully completed a project is a great time to ask a client if they have any friends that may need your services.
## Blogging
Next on the list of organic marketing strategies is blogging. I considered placing blogging at the top of the list because it's such a powerful tool. Over the years I have been contacted by multiple clients that told me they found me via my blog and subsequently hired me.
Blogging is content marketing at its finest. With your blog, you can showcase your skills, position yourself as an expert in the field, and give clients a taste of your personality. If you have never blogged before and are wondering what type of content to write, here are some high-level topics that have worked for me:
* _Tutorials showing how to build features I specialize in._ For example, if you focus on building eCommerce websites, you could write blog posts explaining how to connect an application to a payment gateway.
* _Soft-skill strategies._ For this you can write posts related to experiences you have had while learning how to become a freelancer. An example could be writing about a time when you had a challenging client and how you were able to work with them effectively. These types of posts have multiple benefits since they are effective for content marketing and allow you to contribute to the freelancer community.
## Expert positioning
Blogging is a great way to position yourself as an expert. However, writing blog posts is not the only way to be considered an expert. Another great way to accomplish this feat is to contribute to other blogs and news outlets.
If you look at my personal site you'll see that I've been interviewed or quoted in dozens of blogs and magazines over the years. These outlets include sites such as ReadWriteWeb and the magazine CIO.
And don't worry, you don't have to pay a high-priced PR firm to be quoted on these types of sites. I personally use a service called HARO (<https://www.helpareporter.com/>), which stands for Help A Reporter Out. HARO pairs individuals with reporters around the world. The way it works is that reporters can post requests on HARO asking for interviews or quotes on a topic they are researching. Each day I monitor HARO and when I see a topic that I am familiar with I'll write up a few sentences and send them to the reporter.
Don't let the expert moniker scare you away. There are many times where I am far from an expert in a field that I've been quoted in. However, I'll perform a little time researching a reporter's question and then I'll simply give my opinion. This type of marketing is great because:
* It's free
* Credible reporters are the ones positioning you as an expert
* Many times, the reporter will link back to your website, which helps from an SEO perspective
## Open source contribution
Next on the list of strategies for organically growing a freelance business is contributing to the open source community. When it comes to open source contributions, there are a number of strategies that you can take:
* **Direct code contribution** : This usually comes in the form of creating a code library that other developers can use. One of the more successful Rails development firms in the world, thoughtbot, has taken this approach to the extreme. The thoughtbot team has built libraries such as FactoryGirl, Paperclip, and Administrate.
These Ruby gems are some of the most popular gems in the Ruby development ecosystem and the thoughtbot team released these libraries completely for free. However, I can assure you that the company's open source contributions are directly related to acquiring clients.
* **Tutorials** : If you don't feel that you're quite ready for building a code library that other developers will use, that's perfectly fine. You can find a feature that you feel comfortable developing and you can create a screencast in which you walk through your process for building the component.
* **Contributing to pre-existing libraries** : Another great way to contribute to the open source community is to help add features or fix bugs on pre-existing code libraries. You can easily discover the full list of requested features for a code library by looking at its issue list on GitHub. By taking this approach, you don't have to worry about building a code library from scratch. You can simply add onto another app, which helps the original development team and will give you experience and confidence in working with professional code bases. Personally, I contributed to multiple Eventbrite API RubyGems and built-in functionality that previously didn't exist.
## Social media marketing
No guide that discusses organically growing a freelance business would be complete without mentioning social media marketing. I have to admit that this is probably my least favorite marketing channel. If you peruse Twitter, Facebook, or Instagram it seems like they are cluttered with annoying sales pitches.
However, I have been discovered by multiple clients via my social media accounts. Each day I try to post a development picture on Instagram. And by taking this approach, I have received a number of unsolicited project requests. And several of these requests have turned into freelance clients.
When it comes to social media marketing my recommendation is to find an outlet that you enjoy working with. And once you've picked your favorite channel, put all of your available energy into that specific service.
This is important, because if you pick out an outlet that you don't like, you're not going to want to post on it in a regular basis. And when it comes to social media marketing, consistency is key to success.
# Summary
In summary, when it comes to organically growing a freelance business I focus on three approaches. First, once you have happy clients, work on getting them to refer you to their friends and colleagues. Next, make sure that you're constantly blogging and positioning yourself as an expert in your space. And lastly, find a social media outlet that you enjoy working in and post on it daily.
# Chapter 23. Freelancing Tips – Knowing When to Fire a Client
If you're starting on your freelance journey, or still even considering it, the topic of this chapter may seem insane. However, I can ensure you that knowing when to fire a client is a critical component of building a successful freelance business. It can also sharpen up the same decision-making skills you use every day as a developer in general.
# My urgent client
A few years ago, I was hired by a fast-rising startup. The company had skyrocketing growth, and I was hired to build their platform. After going through the interview process, I was hired. Soon after taking the client, I met with a developer who was leaving the organization.
After the developer walked me through the system he gave me a word of warning. He mentioned that the CEO of the company had a favorite word: **urgent**. I filed the information in the back of my mind and started working on the application. Within days, I learned why the previous developer left the fast-growing startup. Literally EVERY email the CEO sent me contained the word urgent in some form or another.
Through the course of around a year I worked with the company and built out the full system. However, I noticed that my quality of life was negatively affected by this single client. My nights and weekends were no longer filled with spending time with my family. Instead that time was spent working through countless urgent tasks from the client.
## When to fire a client
After a while it dawned on me that if I continued to work with this client, I was actually losing the benefits of being a freelancer. Being a freelancer is supposed to result in freedom and making my own schedule, right?
Once I realized that this client was making my life worse instead of better, I put a plan in place and told him that I would no longer be working with him. Through that experience I developed a system for deciding when to fire a client. There are three criteria that have to be met, and I've listed them here.
### #1 – being treated like an employee
First on the list for deciding when to fire a client is when you are treated like an employee. Being treated like an employee typically means that you're expected to be on the company's schedule and thus limit your own freedom/flexibility. I have had multiple times where a client appeared to forget that I was a freelancer. Some telltale signs of this happening are:
* _When a client is frustrated that they can't communicate with you 24/7_. I once had a client that I had to fire because their employees would send me Skype messages all day and night for trivial issues. And then they would be frustrated if I didn't respond immediately.
* _No organized communication_. Over the years, I've had clients who would let multiple employees message or call me any time they needed a task completed. It's important to have a set number of project stakeholders. If a client lets any/all employees send you requests it's pretty much guaranteed that communication conflicts will occur.
When I realize that a client is treating me like an employee, I'll approach the CEO or whoever my direct report is, and I'll convey my concerns. Many times, this will fix the issue. However, there have been times where the problem persists and I have been forced to fire the client.
### #2 – tyranny of urgent
Returning to my story from the beginning of this chapter, a top reason for knowing when to fire a client is when they can't separate urgent from normal tasks. I remember a time where this specific client set up a project management job board. I kid you not, 90% of the tasks were marked with the tag urgent.
Not only is this a stressful situation, it is also a recipe for failure. When all tasks are marked as urgent it essentially means that none of them are urgent because there's no designation between the projects.
In the book _Rework_ , _Fried_ and _Hansson_ recommend that companies remove the word urgent from their dictionaries. I highly recommend this approach.
Typically, when a client marks tasks as **urgent** or **ASAP** , it means that they don't know how to properly manage a project. And poor project management skills are not an attribute you want in a client, because a client who doesn't know how to manage a project will eventually blame you for not implementing their plan properly.
### #3 – toxic environment
Last on the list for knowing when to fire a client is when it's a toxic environment. I've been fortunate to not run into this situation very often. However, over the years I have a had a few toxic clients. Being toxic can take a number of forms, including:
* Constant negativity
* Poor communication
* Unrealistic deadlines
* Moral/integrity issues
I can think of one client that embodied each of these traits. He hired me to build an application and gave an incredibly vague list of requirements. Throughout the build of the project I would send daily project updates and he would go weeks without giving feedback at all.
When I would hear from him, all his comments were negative. In many of the cases he would be upset for not implementing features that he had never even mentioned in his vague list of features. Needless to say, I fired the client on the spot and moved on to greener pastures.
# The joy of firing a client
When I started out as a freelancer the thought of firing a client seemed crazy. However, as I built up my business, I came to the realization that firing a client that constantly brought stress into my life actually resulted in making me a better freelancer.
Toxic clients are not fun to work with. They cause anxiety and kill the joy that freelancing should bring. So, I'm constantly pruning my client list. And the more experienced I get as a freelancer, the better my client list has become. This has resulted in more joy for me and better performance in regard to what I produce for clients.
# Chapter 24. Dodging Silver Bullets for Scalable Freelance Projects
Does the idea of a code library fulfilling a significant portion of a freelance project that you're contracted into, sound appealing? It's OK, you can be honest and say yes, whether you're an active freelance developer today, or will be in the future.
When I started out on my freelance journey, any time I came across a large feature request, the first thing I'd do was check to see if there was a code library that took care of the requirement. In the development world, code libraries like these are called silver bullets.
However, as appealing as it may sound for a plugin or library to take care of the lion's share of a project, it's been my experience that this approach ends up taking longer to implement than simply building the components from scratch.
# The problem with silver bullets
Let's start off with a practical case study. A few years ago, I took on a fleet management project. The set of requirements included features such as:
* Having full CRUD capabilities for a number of database tables
* Creating a search engine that could search through various attributes of each database table
* The ability to use filters to drill down data
* And a number of other items related to a reporting dashboard
After starting the project, I started to research code libraries that would work as silver bullets and take care of the key features. After a few days, I came across the RubyGem called rails admin. The rails admin gem is pretty impressive. It includes features such as:
* Easily querying database tables
* Implement custom filters for running advanced queries
* Export records to Excel/CSV
* Add new records from the dashboard
* Edit/delete records
As you may have noticed, this code library looks like it fits nicely with the set of requirements for the project I had.
## Silver bullet customization
I quickly went to work building out the application. And I integrated the rails admin gem as a cornerstone component of the project. The client was ecstatic during the demo. They absolutely loved the application and they were shocked I could build the app so quickly.
So, what was the problem? The issues started when the client started to ask for new features. After testing the application out for a few weeks, they came back with feature requests such as:
* Being able to save common queries
* Export out to rare file formats that worked with their accounting system
* Integrate a tax API to calculate depreciation
Each of these new features are relatively common requests for a fleet management system. There was just one problem. Because I built the entire application around the rails admin code library I was limited to the features that the gem offered. And this was a problem because rails admin turned out to be incredibly difficult to modify.
In fact, after several weeks of tedious work, I concluded that it would take less time to build the entire system from scratch as opposed to customize the gem itself. So, the end result was that I wasted quite a bit of time and the project was delayed, all because I thought a single silver bullet code library was going to be able to take care of a significant portion of the application.
## Becoming a sharp shooter with code libraries
So, does this mean that I'm suggesting that you stay away from code libraries entirely? No, not at all! Part of the reason why freelancing is possible (and affordable for clients) is that you don't have to create 100% of an app's functionality from scratch.
Instead, I've learned to be selective about the code libraries that I integrate. Let's look at a couple of the code libraries that I regularly use on projects:
* **Pundit** , for building a permission structure. Pundit is a lightweight code library that is easy to customize and doesn't lock you into a narrow set of permission features.
* **Devise** , for authentication. Writing an authentication system from scratch is time consuming. The Devise gem allows Rails developers to quickly integrate features such as registration, login, logout, and advanced components such as secure password retrieval. Additionally, Devise is customizable and I've rarely run into a situation where it didn't work for an app's requirements.
Do you notice how these code libraries operate like helper libraries for specific features? My rule for integrating code libraries into projects is that I pick out packages that assist with small elements of a project. And I shy away from silver bullets that promise to take care of large portions of a project's functionality, but are difficult to customize.
# Chapter 25. A Freelance Guide to Managing Advanced Features
One issue that every freelancer comes across at some time or another is managing advanced features. And by advanced features I mean that a client asks you to build functionality that you've never built before when this happens to you.
Notice that I said _when_ and not _if_? That's because every freelancer, no matter how experienced, has been asked to build something they've never created before.
# Managing advanced features
Over the years, I've been asked to build a wide range of features. Some of them I had experience with, and others... not so much. A good example of this was a number of years ago when I was asked to build out a GPS tracking iPhone app. The client wanted to track their employees in the field and allow them to remotely submit tickets. Sounds like a great idea, right? I thought so too. There were just a couple of issues:
* At the time, I hadn't built a single mobile application, much less a production app
* My experience with real-time GPS tracking was minimal
Due to my lack of experience, my first thought was to pass on the project. However, the client was a Fortune 500 company and at the time I did have long-term plans on building out my mobile portfolio. Not to mention that the job paid well over $100,000. With these factors in mind I took the offer.
# The talent pool
Now I would have been insane if I thought I could build an application like the one they requested by myself. With zero experience and no domain expertise in mobile apps, the project would have died before it even started. Therefore, as soon as I signed the contract I searched Upwork for a mobile specialist to help build the application.
I found a great iOS developer and we agreed on a mutually beneficial contract where I would manage the client and he would build out the application. This was a foreign concept to me since all the other projects I worked on were ones where I built out 100% of the functionality.
However, if you decide to take on projects that you have limited expertise in, partnering with other developers is a great way to expand into new markets.
# The process
The application development process was different than any I had ever experienced before. I was used to working on every component of a client application. However, for this application I limited my work to the web API development. This allowed me to work with the mobile developer on a daily basis and after around six months we completed the project.
As a controlling person, I found the process challenging on a number of levels. The main issue I ran into was due to feature scheduling. Since the mobile developer was remote, I had to work with his schedule. There were a number of times when this caused issues because I would have an API component completed and I had to wait until he completed a mobile feature, and vice versa.
However, after switching to a Kanban project management board we could schedule our tasks in a more organized manner.
## Kanban
To review Kanban boards, remember that they are a project management system where you have a number of columns designated as:
* Pending
* Working
* Under review
* Completed
We organized each of our tasks into each of these columns. And from that point we were both able to see what features needed to be worked on. Being able to see the stage for each component allowed us to move forward in a more organized fashion than when we were working independently.
# The result
After a number of sleepless nights and few frustrated back and forth emails we finished the project. And not only was the project a success, the application has processed tens of millions of dollars worth of tickets and is still used to this day by one of the world's largest energy companies.
# Summary
So, when a client asks you to build a feature that you have zero experience with, my advice is to partner with a specialist in that area. The example I gave in this chapter was specific to situations where a freelancer has literally no experience building a feature.
There have also been times where I was asked to build a feature I had experience with, but I wasn't comfortable committing to building a production application. In those cases, I hired a mentor to answer questions that I had during the development process, and those projects turned out quite well. And they had the added bonus of teaching me how to build advanced features I had limited experience with.
In summary, when it comes to managing advanced features for clients, don't let your lack of experience stop you from getting jobs. As a freelancer, you have access to a nearly limitless supply of resources to help you build any project.
# A caveat
Before you go out and take on a dozen jobs that you have no clue how to build, let me add a word of caution.
I have seen freelancers and even large software development agencies take the approach of believing that they can simply hire outsourcers to do all their work for them. This approach will fail every time.
To work with outside contractors, you need to work with them daily. Did you notice how one of the prerequisites to the mobile application I worked on in this chapter was that I had a clear communication channel with the mobile developer?
If I would have simply sent him a list of requirements and waited for the finished product the project would have failed miserably. Working with contractors is typically just as time consuming as writing the code yourself. So, don't think that outside contractors are a magic bullet that will do all of your work for you.
# Chapter 26. Freelancer Interviews – Practical Tips for Taking Over a Legacy Application
In this chapter, I interview a freelance developer that I have a tremendous amount of respect for, Derek Harrington. In fact, when I decided to launch DevCamp, I had to let go of a number of my freelance clients. And Derek was who I handed the majority of my clients to. Based on my experience with him over the years, I knew he would take great care of the clients and that they would be pleased with his expertise.
In this chapter, I ask Derek a few questions related to freelancing. Specifically, we discuss practical tips for taking over a legacy application.
Derek's tips can of course apply to any developer whether you're a freelancer or not in your current situation – which is why I recommend for any developer to learn from Derek's freelance approach to this common coding situation!
**What is the first thing you do when you take over a legacy application?**
> Write tests. When you identify the pieces of the code that need refactoring, write specs first to cover the functionality of the feature, then refactor the code and ensure your tests still pass.
**What other practical tips for taking over a legacy application that have worked for you in the past?**
> In evaluating a legacy codebase, use the previous developer if you have access to him/her. Lean on them for info. Ask questions like "what would he do differently?", "what bits of code did he really want to refactor but never got around to it?", "what part of the code is he/she the most proud of?". Lean on their experience to help guide your evaluation. Many times, bad legacy code isn't so much the result of an incompetent developer, but of poor project management and deadline-driven developer pressure.
**What are your thoughts on refactoring a code base Skill Up: A Software Developer's Guide to Life and Career starting over from scratch?**
> Resist the need to rewrite everything so it's perfect. We've all inherited some nasty codebases and if we're going to be honest about it, we've all been that culprit more than once in our careers. But don't rewrite for the sake of rewriting. That's irresponsible.
>
> If you're going to re-write from scratch, make sure you're doing it for the right reasons. We all want to work on brand new projects using the most recent versions of every new technology framework. It's much more enjoyable.
>
> Sometimes it's not the best thing for the project and the client though. Make sure your decision is justified by more than "you wanting to do it". Sloppy code can be cleaned up. Tests can be added. Versions can be upgraded. Old code with poor app architecture riddled with brittle, unstable features and a failure to use any general best practices or third-party tools can be a good justification for rewriting. Lack of understanding of a confusing app is not a justifiable reason.
**When is the best time to work on fixing poorly written code?**
> Just like re-writing an application from scratch, it's irresponsible to leave messy code that every developer on the team is going to touch. There's a time to let bad code be. But not when you're tripping it over it every time you work on the app.
# Chapter 27. Five Tips for Taking Over a Legacy Application
Let`s continue discussing the topic of taking over a legacy application. It's one of the dirty little secrets in the freelance world that a high percentage of the projects that you'll be asked to work on are actually legacy applications, which means that you'll be taking over or working with other developers on pre-existing apps.
There have been a number of times where I've had great experiences taking over a legacy application. Notably I was hired a few years ago to work on a legacy app for Eventbrite, and I was very pleased to find a very well configured codebase. It only took me about a week to become familiar with the inner workings of the application, and I could start building new features right away, it was a great experience.
However, that rarely occurs, typically freelancers are taking over a legacy application because the previous developer was fired from the project or due to the app owner having issues with the performance of the software.
As a case in point, a few years ago I was asked to become the lead developer for a legacy Rails application that had been around for a while and already had multiple developers. This was already a bit of a red flag since well-written applications are typically much easier to maintain, and therefore the original developers are usually still around in some fashion or another. And to put it nicely the app code was convoluted and even after a year it was still difficult to add new features. The legacy code was so fragile that one change could have a domino effect and break other features, with a number of the bugs not showing up until weeks later.
Needless to say, the situation was a mess. I was explaining my predicament to a good friend of mine who was a pretty experienced developer, and he recommended I read _Working Effectively with Legacy Code_ by _Michael Feathers_. Thankfully, I could take what I learned in that book to help completely revamp the application that I had been having issues with. I'll now share with you the tips and techniques I learned there about taking over legacy projects.
# Tips for taking over a legacy application
While there are a number of techniques you need to apply to work with a legacy application, the first should be building a comprehensive test suite.
## Creating a test suite
No matter what language or framework that you work in, you will be able to create automated tests that capture the functionality of the application. So, in the legacy application I was working on I started creating tests for each model.
I began with basic unit tests and then started branching out to integration tests that ensured that the various elements of the codebase were communicating properly with each other. Going through this process had the added bonus that I became more familiar with the structure of the app and I could refactor the code as I implemented the tests.
## Adding new features via TDD
Once the test suite was built, I started building all new features via the **TDD** ( **test-driven development** ) method, which ensured that the test suite was up to date. By utilizing this process, it also made it possible to ensure that the new features that I added wouldn't break pre-existing functionality. This is called regression testing.
## Breaking out specific features into microservices
The further I got into the codebase, I started to notice that the app had become bloated with features, and many of the components didn't need to be included in the core application.
Therefore, I slowly started creating microservice applications that handled isolated pieces of functionality. Some examples were: creating a microservice that managed the user notification system and building an app that processed the reporting engine. After creating the microservices, I could get rid of significant portions of the legacy code and then simply wire up the legacy application with the new microservices so they communicated properly.
## DRY up the codebase
In many legacy applications, you'll run into duplicate code that causes a number of problems, including the issue of having to make one change in multiple places in the codebase.
An example of this was how the application I was working on dealt with view templates. There were a number of view files with identical HTML code. I could refactor these components into partials that could be shared across the application, which allowed me to make a single code change that would populate throughout the app.
The topic of taking over a legacy application is important to understand, not only for the reason of being prepared for what steps you need to take to work on a legacy app, but also so you will have a better idea of how to build applications from scratch.
Remembering the Eventbrite application that I mentioned earlier, that application was easy to work with and add features to because it had been built from day one using each of the techniques mentioned in this chapter.
If you develop an application from scratch using these best of breed techniques, you will make it easier on yourself when you're adding features in the future. It will have the added benefit that any new developers that may work on the application in the future will be able to start adding new features easily and they'll appreciate the extra work you put into the development process.
# Summary
I hope that this has been a helpful guide for taking over a legacy application and that you can apply it on the projects that you're working on.
# Chapter 28. Guide to Freelancing – Starting Over Versus Refactoring
As a freelancer or as part of an organization, you will come across many times where you make a decision on starting over versus refactoring on a legacy project. Over the years I have come across this issue more times than I can count.
# The legacy scenario
Typically, the situation sounds something like this. I'll get hired by a client who has a legacy application. The application is usually a few years old and has been managed by a number of developers. I've discovered that usually the code project started out small, and it grew from there.
Somewhere along the way the application lost its way. Instead of using a scalable application design approach, the previous developer patched new features on and the codebase devolved into a convoluted mess. Eventually, every new feature causes another component to break and the client gets so frustrated he decides to hire me.
Does this scenario sound familiar to you at all? As a freelancer, I've been on both sides of the legacy code spectrum. When I was a new developer I built new projects that got out of hand and I lost the clients. And as I matured as a developer I started getting hired to take over legacy projects.
I can tell you from experience that neither side of this scenario are fun, especially when the client brings up the dreaded topic of starting over versus refactoring.
# Starting over versus refactoring
When I was a young and naive freelancer I dreaded the idea of re-factoring a legacy application. The second that the client brought up the possibility of starting over from scratch, I jumped on it!
However, the more projects I work on, the more my mindset has changed. Over the years I've put together a system to help me decide between starting over versus refactoring a project. And that's what I'm going to walk through here. The steps I follow are:
1. Removing the fear factor.
2. Analyzing the 80/20 principle.
3. Building an automated bug list.
4. Becoming the client.
## #1 – removing the fear factor
One of the reasons why inexperienced freelancers tend to opt for starting a project over is because of fear. And fear is rarely a good reason to make any decision (unless you're running away from a wild animal or something like that). So, before I make a decision, the first task I perform is taking fear out of the equation. I'll ask myself:
_"If you weren't afraid of the unknown issues with this codebase, what would your decision be?"_
Once fear has been removed I can look at the project from an unbiased viewpoint.
## #2 – analyzing the 80/20 principle
The 80/20 principle has a number of practical ramifications. You may have heard it being used to say that 20% of the people make 80% of the wealth. Or that 20% of a customer base generates 80% of a company's revenue. However, I've also seen that the 80/20 principle can work well for deciding between starting over versus refactoring.
Too many times I've seen a freelancer start a project from scratch when the legacy application already contained 80% of the functionality needed from the client. This means that the developer only needed to take care of the remaining 20%.
If you look at the numbers the answer becomes readily apparent. Would you rather perform 100% of the work (which is what would be needed when starting from scratch), or only 20%?
So, in this step I take a step back and I analyze what features the client is asking me to build. If the legacy application is functional and simply has a messy codebase, it's rarely the smart move to start over from scratch. Typically, in this case I'll add the new feature and then start refactoring the application one module at a time.
## #3 – building an automated bug list
Moving down my list I'll next use automated tools for analyzing the application. Pretty much every programming language and framework has a wide range of analysis tools. I'll utilize these tools to generate a set of issues for the legacy application. I like this step because it accomplishes two key goals:
1. It gives me a practical strategy for what needs to be fixed in the application.
2. Since it's automated, these types of tools are unbiased. Your personal judgments on the previous developer's coding style are taken out of the equation. And the focus is centered solely around the project itself.
While each project is unique, the tests I run usually focus on giving me a report on:
* Potential security issues
* Best practices
* Code that is not being utilized (this is vital!)
## #4 – becoming the client
Lastly, I try my best to remove my personal feelings from the decision. Instead, I focus on taking the perspective of the client. I'll ask myself:
_"If I was a fully informed client, would I really want to pay for a developer to start the project over from scratch?"_
Using some fuzzy math, I'd estimate that around 9/10 times my answer to this question is that an informed client would request a refactor over starting over. Starting a project over is expensive, and there's no guarantee that the new codebase is going to be perfect.
In fact, I can pretty much guarantee that there will be issues with a brand-new application. I've seen multiple times where a legacy application was replaced by a new piece of software that had the same number of bugs.
# When should you start over?
So far, this chapter has heavily favored refactoring an application. However, there are times when starting over from scratch is a better approach. Some of the rationales that make starting over a wise decision for the client are:
* **A complete change in architecture** : I have had clients who requested that I migrate an application from being a monolith (a large single application) to becoming microservice-based (a number of applications that each perform a single feature). In cases like this, it wouldn't make any sense to try to keep the legacy application since the core application structure would have to change.
* **Moving to a different language/framework** : Over the years, I've had a number of clients who had old ASP .NET and PHP applications approach me to rebuild their systems in Ruby. When it comes to changes languages or web frameworks it wouldn't be possible to retain the legacy application.
Notice how both these key reasons had nothing to do with bugs or messy code? The only time I'd recommend for a client to start over from scratch is if it's literally impossible to retain the legacy application.
# Summary
In summary, the key to remember is to go through the system of checks whenever you're asked to decide between starting over versus refactoring. The more experienced you get as a developer, the more you'll realize that it's incredibly rare that a functioning legacy application cannot be saved.
# Chapter 29. Should You Use TDD on Freelance Projects? – Comparing Quality Versus Speed
When approaching a new freelance project, one of the first questions many developers and designers ask is:
_Do you want it done fast... or properly?_
In this chapter, I'm going to specifically discuss if you should use TDD on freelance projects. However, the concepts I'll discuss now apply to any type of quality control system. So, this is great advice for any developer.
# Quality versus Speed
There's an old software engineering rule that states that there are three options you have when building a project:
* Quality
* Speed
* Cost
This diagram represents the principle as a Venn diagram. The elusive center is where you have a project that's built quickly, for a low price, and was created with industry-wide best practices. In my experience, it's rare for a project to reside inside of this sweet spot. Instead, I'll tell clients that they can pick two out of the three. For example:
* You can have the project built properly for a low cost, but it's going to take quite a while to develop. This is because the project will need to sit on the back burner since higher paying projects will need to take precedence.
* Alternatively, the project can be built quickly at a low cost. However, this approach won't allow for the time needed to follow best practices, such as building automated tests into the application. I rarely offer this option to clients because it's too tempting for them, and I've seen from experience that these projects always end badly.
# TDD on freelance projects
In this chapter, I've selected the concept of **test-driven development** ( **TDD** ) as a measurement of project quality for one key reason. Every time that I've been handed a messy legacy project to work on, there is always one common characteristic that they share:
_The code doesn't have a comprehensive test suite._
On the other hand, whenever I start working on a high-quality application I've discovered that these projects pretty much always have solid test coverage.
So, I'm not saying that a full test suite is required for a project to be considered a high-quality product. However, in my experience tests seem to be a key indicator that determines how well an application was built.
## Making the decision
When you are embarking on a new project how should you decide if you should use TDD?
### Giving no choice
I know plenty of developers who simply do not give clients a choice in the matter. All the code that they write will be tested, period and full stop. This helps make the decision process more straightforward. This is the approach I take now, but that's only because I now have the ability to be more picky when it comes to the clients I take on.
However, if you are new to freelancing and you need clients, it can be difficult to tell a client that a project will be around double the time and cost. If you don't have a strong set of pre-existence, you may find yourself in a situation where you price yourself out of the market.
### Letting the client decide
Alternatively, you can let the client decide on what approach they want you to take. In this situation, you propose the pros and cons to building a full test suite compared with only building the application itself. If you have an intelligent client they will most likely see the benefits of including tests and choose for the pricier option.
This is an effective strategy because it allows for you to bring the client into the decision-making process, which will make them feel involved in the work. And if the client is still looking at other freelancers, this approach may help win him over.
Another benefit to letting the client decide is that his response may give you insight into how he thinks. If he acts like tests are a pointless luxury and says that he simply cares about getting the project complete, he might be a nightmare client. And in cases like that you are better off moving onto more informed people to work for.
### Using common sense
Lastly, make sure that you're using common sense. Imagine being asked to build out a simple corporate website. In cases like this you only need to write some basic tests. At the most this should only add an hour or so to the project.
There is no need to bog down the process writing tests that verify that every CSS class and ID are shown on the page. As with most concepts in freelancing and life, common sense is one of your greatest tools.
# Chapter 30. Automating Client Updates as a Freelance Developer
If you have limited freelancing experience, it may surprise you to discover that a significant portion of a developer's day is spent detailing the work performed for that day. In this chapter, I'm going to walk through automating client updates so that you can be as efficient as possible.
# Importance of daily updates
Before diving into how we can automate updates to clients, let's discuss what a proper update is and what it entails. An update is a message sent to a client, usually every day or at least every day that you're working on the client's project. The days are long gone where clients would hire a freelancer and the developer would disappear for a few months until they brought back a finished product.
Nowadays, clients want to have a transparent view of the work performed. This is especially true if a client is paying you on an hourly basis. This makes sense because if you hired someone and paid them for their time, wouldn't you want to know how the time was spent?
Regular and explicit updates are also an important way that you can distinguish yourself from offshore development teams. Over the years I've worked with development teams across the world. And the number one issue I constantly had with them was finding out what they did each day. So, if you can give a transparent view into the work that you perform for a client, it can give you an edge over cheaper, offshore freelancers.
## An example of client update
So, what does a good daily update look like? Here is one I took from a real-world client update:
* Integrated CSS fix for the location widget
* Continued working on bug fix for the well on the right side of the page
* Updated CSS for the locations widget on the city pages
* Integrated the checker for posts on the city-specific show pages
* Updated sign up buttons
* Temporarily hide sponsor text
* Implemented changes to the contact us text
* Implemented custom sub division with master division annotation for the forms
* Updated the edit label on the post show view
Notice how these updates are practical and informative. None of the items are too technical, since overly technical updates would simply confuse clients.
## Automating client updates
So, we've established that client updates are important and we know what a good client update looks like. However, if we have to type these updates in from scratch every day for multiple clients, it would tally up to quite a bit of time. I'm not a fan of wasting time and I doubt you are either, which poses a dilemma:
1. We need to create a detailed updates of all the work we do each day.
2. But we don't want to waste time writing reports (and isn't our hatred of writing boring reports what made us want to become freelancers in the first place?!).
Whenever I come across a situation like this, I try to see if there is any way that I can automate a boring task. Thankfully there is.
## Version control to the rescue
In the beginning, I was writing out all the daily updates manually. However, if I wrote them at the end of the day, I had to go back through all my GitHub commit messages to reference the work that I did.
After going through this process for a while, it dawned on me that if I simply added a little more detail to my GitHub updates, I could simply copy and paste them each day and I wouldn't have to write them from scratch again.
This process ended up saving me a considerable amount of time each day because for best practice reasons, I had already been writing GitHub commit messages for each new feature I implemented. So, now I can remove the duplicate work I had been doing.
Here is a screenshot of a GitHub project where I used the technique of leveraging commit messages for automating client updates:
# Summary
I hope that this has been a helpful guide for automating client updates, and that you can leverage it with your own clients.
# Chapter 31. Freelance Requirement Elicitation – A Guide for Feature Development
Imagine for a minute that you're a freelance developer who was handed a new feature to build by a client. Then picture yourself building an elegant feature, all the code working perfectly. You follow best practices and ensure that all the potential edge case scenarios are covered.
Now imagine that you're demoing the bright and shiny new feature to the client. But instead of telling you that you're the best developer in the world and they're going to name their first child after you, they look at the application confused, because what you built didn't match what they had in their mind at all.
This is a scenario that is played out all too often in the freelance development world. And in many cases, it's due to a poor requirement elicitation process. The story I just mentioned is not a made-up parable, it happened to me recently. And when I say recently, I mean yesterday (at the time I wrote this).
_My Name is Jordan and I Wrote a Poor Requirements Doc... "Hi Jordan..."_
So, what did I do wrong? The issue was caused by me rushing through the requirement elicitation phase. I have worked for this specific client for over 5 years and I got lazy confirming the exact set of requirements needed for the feature.
# Freelance requirement elicitation
Let's walk through what happened so you can avoid the same embarrassment and wasted time.
## How it started
A few weeks ago, the client contacted me and said that an application I built for them needed a new feature. The application is an invoicing system that their drivers utilize to generate invoices for clients:
In an email, the client attached this spreadsheet. He said that the application had to generate this invoice to give to the customer.
## The build
After receiving the email, I spent a few days modeling the new feature. I put a list of all the messages that would be passed between modules. I built UML diagrams to ensure the data was modeled properly. After careful planning, I spent two weeks building the new feature and it came out perfectly.
To be 100% honest, I was very proud of the work that I did. The feature was flawless and completely bug free. It also fit in perfectly with the rest of the application. I deployed the code to the staging server and I waited for the client to start showering me with praise... but the praise never happened.
## The problem
I emailed the client and gave a video demo of the feature. A few hours later I received an email from the client that said:
> I'm confused, what exactly is all of this? In my email, I just meant that we need the invoices to be formatted like this spreadsheet.
So, it turned out that the client didn't want a new module built into the application at all. Instead, they simply wanted an additional format option for their invoices.
## Who was at fault?
So, who exactly was at fault? It may seem natural to put the blame on the client since they didn't make their request clear at all. And I was tempted to get upset and blame them (especially for the first 10-20 seconds of my fury). But then I realized that this issue was completely within my control.
As freelancers, it's our job to manage each stage of a project. If we rush through the requirement elicitation phase, anything that happens after that stage will fall on us.
# A better way
So how could this have been avoided? Let's walk through the process I should have followed and that would have led to a better outcome for myself and the client.
## Step 1
Right after getting the email I should have responded to the client with clarification questions. Examples might have been:
* _Do you want this to be on a new page of the application?_ This is better than saying something like: _Do you want this to be a new module?_ Because a nontechnical client isn't going to know what a module is. But they will understand what a new page on the site is.
* _How will this interact with other parts of the website?_ This question would have instantly given me the feedback to know that this spreadsheet was simply meant to be a different invoice formatting option.
* _Can you describe the flow of how this will be generated?_ This is one of my favorite questions to ask because it forces the client to be explicit with how a new feature should work. Many times, I'll ask a client to create a PowerPoint slide deck showing the flow they want from a feature.
## Step 2
After asking clarification questions, I should have followed up with a prototype. I could use a tool such as InVision or even a simple PowerPoint deck where each slide held a different page of the proposed new feature. Examples would be:
* Starting with slide 1, this is where you can click on a button to get to the new page.
* On slide 2 I'd show the form page where the user would enter the information.
* Lastly, on slide 3 I would show the invoice that was generated by the new feature.
# A better ending
If I would have followed these two steps, it would have taken me anywhere from few minutes to hours to establish what feature was actually needed. As you can imagine, this is a much better option compared to wasting weeks of development time.
# Summary
I hope that this has been a helpful chapter to freelance requirement elicitation and that you'll be able to learn from my mistake and apply it to your own business.
# Chapter 32. How to Remotely Demo Work for Freelance Clients?
So, you have a freelance client and you're ready to show off your work, but how can you showcase a project when you work remotely? If the entire application is completed, you could simply send the client a link to test it out. However, it's been my experience that this approach is not a great idea. As the designer or developer, you know the inner workings of the app.
The client, on the other hand, especially if they're of the non-technical variety, will stumble through testing the application out. Even if you did a great job on the project, if the client doesn't know how to use the software they're not going to be happy with your work.
# Why proper demonstrations are important
Over the years, I've discovered that well thought-out demonstrations are a key to successful projects. There are two main reasons for this:
1. A proper demo will let you control the flow of the app. You can control the focus of the work and spend time showcasing how the system works. This will essentially function as a tutorial for the client of the app. A well-organized demo will educate the client on the application and remove many of the common issues related to user experience confusion.
2. You will get practical feedback. No matter how well you think you understood the client requirements, there will always be misunderstandings, especially early in the development process.
Thankfully, if you put together a proper demonstration of the software you can get a clearer view of the client's vision. If you simply sent off a link to the app for the client to test, many of the feedback items they would send back would be related to not understanding how the system functions. However, if you can clearly show how the app works, it will shortcut this process and let you understand the actual fixes sooner.
# Review of services to remotely demo work
There are a number of ways that you can remotely demo work for clients. I'm going to go through the processes I've used throughout the past few years and discuss when each option is ideal.
## Screencast
The first option I'll review is creating a screencast. This is a great option if there are a number of clients that will need to review the feature:
At its core, a screencast is you recording yourself walking through the application. If it's a web-based application, you can have the video showcase each feature of the application. In my own work, I use tools such as:
* Zooming in to specific screen zones
* I highlight sections of the screen that I want to focus on
This process only takes a short period of time. However, it is well worth it because it enables clients to have a tangible walk-through of the system.
You can use a number of tools for recording screencasts. My personal favorite is Camtasia; however, it is a little expensive. So, if you are trying to keep a low budget, there are countless free options, such as Screencast-O-Matic.
After you finish filming a screencast demo, you can upload it to YouTube, Vimeo, or a video hosting site. From there, you can have the client view the link at their convenience.
## A remote desktop
Next on the list of tools to remotely demo work is remote desktop sessions. Every client is slightly different. Many clients are fine with email and video demonstrations. However, other clients want more of a personal touch:
If a client likes to have full interaction with you during the demo, a remote desktop tool may be the best choice for showcasing your work. I have also found that this option is ideal when I'm working for other developers. This is mainly because remote desktop demos allow for pair programming. There are a few different options when it comes to remote desktop tools:
* **GoToMeeting** : If your client wants to have a traditional live demonstration, services such as GoToMeeting or Join.me work nicely. They will let clients view your screen and they also come with dedicated conference call lines if you're working with multiple stakeholders.
* **Screen sharing** : There are times when you need the ability for you and a client to simultaneously walk through a demo. This is for the scenario I just mentioned where you're working for a developer and he wants to be able to go through the app at the same time as you. My favorite service for this type of demo is Screenhero. It offers an easy way to have multiple users control a screen at the same time and it's pretty affordable.
* **Free options** : If you're on a budget, there are a number of free screen sharing and remote desktop options. Some of the notable ones are Google Hangouts, TeamViewer, and Skype.
## PowerPoint
Last on the list of tools to remotely demo work are PowerPoint-type presentations. I say type because you don't have to actually use PowerPoint. I've used PowerPoint, Google Slides, and Keynote for product demos. I like using slide-based demos early in the project development process.
There are many times where I've built backend functionality that I wanted to confirm was configured properly. However, I realized that if the client saw an ugly user interface, they wouldn't be able to appreciate the work that was performed. So, I took a play out of the Google Venture's playbook and I created a Keynote slide deck. I designed the deck to mimic the user interface that I planned on eventually adding.
From there, I simply loaded the slides with the behavior I had built into the actual application. This approach worked quite well and the client was happy. More importantly, this option let the client focus on what I had built as opposed to requirements that were still on the to-do list.
This option also works quite well for mobile app demonstrations. Mobile apps are pretty complex for clients to demo on their phones. So, a slide-based approach makes it possible to show an app's design and behavior in a more efficient manner.
# Summary
I hope that this has been a helpful freelance guide and will help you remotely demo work for clients.
# Chapter 33. Defining Project Success as a Freelance Developer
When it comes to freelancing, _defining project success_ is a surprisingly challenging task.
# A clear end
Imagine that you're competing in a marathon. When do you know it's the right time to stop running? For me, it's when I cross the finish line. It seems borderline insane to picture running a race without knowing where the finish line is, so why do so many freelancers work on projects without a clear concept of completion? If you don't establish mutually agreed upon project completion criteria with a client, you may find yourself subject to scope creep.
## What is scope creep?
Scope creep in a project is when a client asks for changes in the application that exceed the original set of features. Many times, the client doesn't do this on purpose. A normal progression is for a client to see the development progress and then realize that they forgot a "key" feature.
## When scope creep isn't scope creep
There are times when the right thing to do is incorporate the feature they're asking for. I can think of examples where the client hadn't listed a specific feature, but the feature was truly necessary and was required by pure common sense.
Recently, I headed up the development of an iOS project where the client didn't specify that a push notification needed to be directed to a post. After the application was completed, the client was frustrated that the system didn't have dynamic and clickable push notifications. I could have pointed to the fact that they never asked for the feature. However, in my mind the behavior was a common-sense feature, so I had it added for no extra cost to the client.
## When scope creep goes badly
Scope creep is rarely that easy. You'll discover that typically clients will come up with new ideas and then try to get you to implement them for free. If you haven't established a clear definition for project success, you'll end up with an angry client who thinks that you're trying to overcharge him. Remember that in the client's mind they may not realize that they're asking for a feature outside of the original set of requirements. There are two ways for defining project success. We'll walk through both of them.
### Based on requirements
First and foremost is the traditional approach, which is based on a set of requirements. This approach is OK; however, it rarely works in the real world. This process goes through the following workflow:
1. Write out a comprehensive set of project requirements.
2. The requirements sound something like: user should be able to log in.
3. Each feature has its own requirement.
4. Once all the requirements are implemented, the project is considered complete.
Theoretically, this seems like a great plan. However, in real-world projects, it rarely works. The issue is mainly that even the most experienced developer or project manager won't be able to list every... single... little feature. What will happen is that features will be missed and either you or the client is going to have to compromise to get the project completed.
### Based on a story
So, if defining project success based on requirements isn't practical, what's a better approach? Personally, I have had the most success by building easy-to-follow application stories. What is an application story? Let's take a look at one I wrote for a recent project:
_When an admin user logs into the application she will be shown a custom dashboard that renders all the projects that she manages. From there she can edit project details. She also can navigate to the resource section, user management dashboard, and user audit log._
Notice how a story is different from a set of requirements? When clients are presented with stories it is easier for them to visualize the final product. This leads to them supplying you with the full set of required behavior in the beginning, instead of at the end. Lastly, well-constructed user stories give you a clear definition of project success. Are the stories functioning properly? Then the project is completed, it's that easy.
### The sign off
After the client has approved the full set of application stories, make sure to get a formal sign off from the client. Typically, this means having them sign a document that contains all the stories. This provides a practical agreement that you can point to when all the features have been implemented.
# Summary
I hope that this has been a helpful guide for defining project success as a freelancer and that you can use this approach on your next project.
# Chapter 34. Top Project Management Tools for Freelancers
In this chapter, I'm going to discuss the top project management tools I've used on coding projects. In preparing for this chapter, I went through some directories that listed all the known project management software applications on the market.
To be 100% honest, I was a bit shocked. There are literally thousands of project management options available to freelancers. Thankfully, I've been able to work for a number of clients and worked with various project management apps over the years.
For this chapter, I want to give you a list of the top project management tools along with their respective strengths and weaknesses. With this knowledge, you can decide on which one is the best fit for your project and freelance business.
The following are six of the top project management tools that I've used. I've probably used around a dozen tools; however, I only wanted to list applications that:
* I had actually used on production projects
* I had a good experience with and that I'd recommend to others
# Top project management tools
We'll now look at each tool one by one.
## Basecamp
My favorite application for project management is Basecamp. I'm probably partial to it since its founder, David Heinemeier Hansen, also happens to be the creator of the Ruby on Rails framework, which I use daily:
With that being said, Basecamp has a clean interface focused around to-dos and messaging. Here is a set of the to-dos assigned to me and various DevCamp team members right now:
In addition to project management, I also use Basecamp as my daily planner. Essentially, I move each of my to-dos from one day to another. I like how I can have a number of my projects and their respective to-do lists all shown on one page, as shown here:
And for when I'm wanting a filtered list of what is assigned to myself, I can see only the projects that I'm tagged in as the owner:
For the negatives of using Basecamp, depending on your prior experience with project management software, Basecamp may not feel very natural. If you're used to dashboards based on Gantt charts and similar tracking mechanisms, Basecamp will take some getting used to.
It does have the ability to have all those features via their add-on module. However, at its core Basecamp focuses more on messaging between team members, to-do lists, and scheduling.
Basecamp also doesn't have the best suite of mobile options, I have its iOS app on my phone. However, it's not the most intuitive, especially when compared with a few of its competitors such as Trello.
With that being said, Basecamp is still my go-to choice when it comes to project management software due to its simplicity, speed, and because I'm drawn to its to-do list structure.
## Trello
Another great tool that I've used on a large number of projects is Trello. Trello utilizes a Kanban style of project management. Kanban is a workflow popularized by lean manufacturing proponents, and at a high level it uses the concept of moving cards through different stages of a project's development until they're complete:
Trello essentially gives you a virtual Kanban board and lets you move tasks through each stage of the product development life cycle. For example, here in the image I would move a task from being a **To-Do** to being **Assigned** , to **Working** , to **Under Review** , and finally to **Finished** :
This approach makes it straightforward to visualize each task that's assigned to you, and it also has a good interface for managing a team of developers.
I also like how you can add images and attachments to tasks, which lets you encapsulate all of a task's conversations, data, and statuses in a single screen. Another bonus that Trello offers is that it works nicely on smartphones and tablets, so it's easy to track the progress of applications when you're not at your desk.
Trello has worked well for me on small-to-medium sized projects, but for large projects or complex applications I'll usually opt for Basecamp or the next one on the list: LeanKit.
## LeanKit
When I was managing an entire IT organization, LeanKit was the software I went with for tracking the projects that were being worked on. It doesn't have the attractive user interface that Trello offers and it takes longer to learn.
However, it worked quite well for me when I was managing very large-scale projects, such as enterprise application rollouts to 800+ employees.
Much like Trello, LeanKit utilizes the Kanban strategy of project management. However, it focuses more on enterprises compared with smaller teams. The mobile and tablet applications offered by LeanKit were intuitive and helped me to manage projects even when I was traveling, which was a nice bonus.
LeanKit's strength is in how well it allows you to nest and organize projects and subprojects. After you've learned how the system works, it's relatively straightforward to manage large teams.
One of the biggest negatives of LeanKit was something that I considered very odd. They have a pretty archaic method for payment that requires quite a bit of manual work. I remember times where I was forced to contact the company via email simply to add new users to our account plan. This wasn't a deal killer for me. However, it does make it a poor choice for freelancers who want to work multiple projects from the same account.
## ProWorkflow
ProWorkflow is one of the more standard project management options on this list of top project management tools. It does a good job of combining features such as timelines, task management, and working with teams. I only worked one project that utilized ProWorkflow. However, it was a good experience and I didn't have any complaints using the software.
## Wrike
I have a bit of a love/hate relationship with Wrike as a project management tool. If you're managing a team of developers, Wrike is a great application to use. As a project manager, you're able to use it to see the status of each project along with detailed analytics for every task that is being worked on.
However, if you are a developer working on Wrike, there is a pretty steep learning curve. I would like to think that I'm pretty adept at understanding how to use a new piece of software. However, I kept getting lost in Wrike's dashboard and had a difficult time finding where to post updates to clients. Eventually, I asked the client to go through a screen sharing session where we walked through each dashboard that I would use.
With all that being said, Wrike is a good option if you're managing a team of freelancers on multiple projects, and that's why I put it on this list.
## GitHub
This may seem like an odd option to put on a list of top project management tools since technically, GitHub doesn't market itself as project management software.
However, when I'm working on a project that only has developers, I've found that using GitHub's issue tracking module doubles as a project management tool:
And it makes sense since some of the largest frameworks and languages are open source projects that base all the tasks on issues and features that can be easily tracked on GitHub:
As you can see in this screenshot of a specific issue, you're able to:
* Create issues, which can also be features
* Assign users to each task
* Add links and images via the Markdown syntax
* Mark the issue as being completed
Additionally, with an experienced development team, using GitHub to manage a project has the added benefit of making task management efficient. For example, developers can close tasks automatically based on Git commits. And the easier you make a piece of software to use, the more development teams are going to like it.
# Summary
I hope that this has been a helpful list of the top project management tools that you can use in your development projects and that it will help you decide on what software to use on your next project!
# Chapter 35. Top Freelance Bookkeeping Options for Developers
In this chapter, I'm going to review the top freelance bookkeeping options. Keeping track of finances is probably one of my least favorite parts of being a freelancer. To properly manage a freelance business, you have to manage a wide range of accounting components:
* Profit and loss reports to capture your overall profitability
* Aging reports to see how long it takes to get paid
* Expense logging to ensure you capture all potential deductions
* Accounts receivable system so that you can get paid
* Project tracking tools if you're working on large-scale projects for clients
# Freelance bookkeeping options
I've used a number of bookkeeping solutions over the years, and the following are some of the ones that I've had the best experience with.
## FreshBooks
FreshBooks is the system that I use for my freelance business. As far as accounting software goes, it covers 100% of the requirements I have:
Thankfully, it is also easy to use, which means that I didn't have to spend time learning how to use the software. It has an intuitive interface and allows me to log in, perform whatever tasks I have to, and then get back to working.
### How it works
Here's an important screen for me in FreshBooks:
On this page, I can:
* Select a client.
* Add additional accounting information. This includes items such as a purchase order number.
* Add items to the invoice. I also like how easy it is to add fractional quantity units. If I worked a little over 14 hours, I can enter 14.1 hours and FreshBooks calculates the amount.
* Enter the payment method. This portion of the system makes it easy to let clients know if I want to be paid via standard PayPal, PayPal Business, or via FreshBooks's payment system.
* Add comments. You can post any additional information that will be sent to the client.
### FreshBooks additional features
It would take hours to go through the full system, so I will simply gloss over some of the other features I find the most useful:
* **Invoice sending flexibility** : As great as it is to send invoices electronically, I still have a few clients who prefer paper-based invoices. Thankfully, not only does FreshBooks allow me to print out invoices, they actually mail them for me directly to the client.
* **Expense tracking** : Tracking business expenses is a tedious exercise. However, with the FreshBooks mobile app I can take pictures from my phone, enter the expense details, and the expense will be logged into the system.
* **Creating estimates** : In the past, I would create an estimate in Excel or Word, and then send it to the client. If they approved it I would then have to enter the details into an invoice. However, with FreshBooks I can create an estimate and email that directly to the client. If they decide to move forward with the project, the estimate will automatically transfer into an invoice.
### Weaknesses
As much as I love the FreshBooks system, it does have a few weaknesses. First and foremost, it can get a little expensive if you get a large number of clients. Also, if you start hiring a large number of employees it can become a little unwieldy to manage. Personally, I'd recommend using Freshbooks for any business with under 20 employees. When you grow beyond that point, you'll want to move to a more scalable system.
## QuickBooks
If you've grown out of FreshBooks, QuickBooks may be a good option for our business. QuickBooks has been the industry leader for small business accounting software as long as I've been around. Even though the company has acquired the reputation for being difficult to manage, over the past few years they've done a good job in making the system more flexible.
When you list all the potential features needed by accounting software for a small business, QuickBooks has it all. Also, due to its popularity, there is a good chance that any admins that you hire will already have experience using the system.
With all that being said, I personally wouldn't choose to use QuickBooks for DevCamp or any of the companies I work with. The main reason is because I've seen too many times where companies grow out of QuickBooks and found it very hard to migrate to a new system. I also don't like the reporting engine that the software uses. For example, if you plan on running your financial data through a big data analysis reporting engine, QuickBooks makes it difficult to export it in a format that's easy to use.
## NetSuite
If your business is growing, both in clients and employees, NetSuite is a great bookkeeping option. Technically, NetSuite is probably overkill for the typical freelance business. However, if your development shop starts to turn into a full-fledged digital agency you'll need a robust ERP system. ERP systems are different from traditional bookkeeping software, they'll offer tools such as:
* **Resource planning** : This means you can allocate developer time on a project basis.
* **Advanced tax planning tools** : Once your business hits a certain size, it's important to ensure you're taking advantage of all the potential tax deductions available. Tools such as an ERP system do this for you.
* **Payroll integrations** : Being able to manage your employees, track turnover, and tasks such as that get important as your business grows. However, they're hard to track manually. A system such as NetSuite calculates all your employee tracking metrics so you can use them to manage your team.
# Summary
I hope that this has been a helpful set of freelance bookkeeping options that will help you decide on the right system for your business.
# Chapter 36. Learning the Secret to Get New Clients as a Freelancer
If you're wanting to start a freelance business, the most pressing challenges typically revolve around getting new clients. When I initially launched my freelance business, I struggled to acquire customers.
However, after a few months of trudging through the traditional channels I discovered a great solution that resulted in generating over $290,000 in revenue last year. And it's what I want to discuss in this chapter.
# Where to find new clients
Let's begin by reviewing the list of options for where you can find new clients as a freelancer. I've read countless blog posts and a number of books on the subject, and the following were the most popular recommendations:
* **Friends and family** : This may seem like a natural place to start; however it's been my experience that friends and family typically expect you to work for free or incredibly cheap. This channel can be good if you're just starting out and you need to build a portfolio; however, it is not scalable and usually not too profitable.
* **Network events** : For networking, you can join your local chamber of commerce or find networking groups where you can promote your business. This approach can work well in some cases. However, each time I've tried it I've discovered that there were already several other developers attending working on getting new clients for their own freelance businesses.
I wouldn't let my experience with this option stop you from trying it, but make sure you're prepared to compete with others. I remember attending a chamber of commerce "meet and greet" a few years ago. When everyone around the room introduced themselves and their business, there were half a dozen freelance developers who were all offering the same services.
* Outsourcing services. This is the option that I want to focus on here. Interestingly enough, I was told by a number of freelance "experts" that this avenue was too difficult. However, I found a great way to use it for getting new clients as a freelancer.
# The challenge in getting new clients with outsourcing services
As I have already mentioned, I had a difficult time in the beginning getting new clients. Using services such as oDesk and Elance (which have now merged and are now Upwork) were especially challenging. Even though I had a solid portfolio and a decade of experience, I couldn't get a single client. Some of the challenges were the following:
1. I didn't have any ratings or reviews. Not many clients are willing to take a risk on a developer without some type of recommendation from others.
2. I was priced higher than the majority of other freelancers. The majority of the freelance teams marketing services on outsourcing sites are offshore. This meant that I was having to compete against developers offering to work for, at times, 90% cheaper than my rate. I charge $100 per hour, while the majority of offshore teams are charging $10-$20 per hour.
3. I didn't have the time to pitch each potential client. In regard to marketing my freelance services, I was very streaky. I would get motivated for a few days and send out a large number of pitches. And then I would get depressed that I wasn't getting any replies and I wouldn't send any proposals for weeks.
After struggling for a few months, I knew I didn't have any control over challenges #1 or #2. However, I could do something about #3. I was working a full-time job at the time, while also attending grad school, so my time was very limited. With that in mind I came up with a marketing system. And it actually worked!
# Getting new clients as a freelancer
Since I'm a little bit on the stubborn side it took me a while to admit it, but I finally came to terms with the fact that I wasn't getting new clients as a freelancer, especially with the methods that I'd been trying up to that time. So, I put a plan into action that involved hiring some freelancers of my own. I assembled a team that helped fill in my weak areas.
To start getting new clients as a freelancer, I knew I had to have clearly written proposals that described my services. And I also knew that the proposals would have to be sent out 24/7.
## Proposal material
With that in mind I researched sales copy writers on Upwork and hired a talented marketer who created three different proposals that I could use.
This included sales copy that advertised my experience, portfolio, and expertise as a developer. I had three versions created because I wanted each one to be targeted to a specific type of project. For example, one of the proposals focused on enterprise projects. Another proposal was targeted at building APIs, while the third had content geared toward startups.
## Sending out constant proposals
With a full set of professional proposals, I was ready to implement the second step of my plan: consistently sending out proposals. For this, I hired a detail-oriented and fluent virtual assistant from the Philippines named Sy.
I could hire Sy for $6.50 an hour and he paid for himself in the first week! I walked him through the proposals and described the types of projects that I wanted to get hired for.
After I was confident that Sy clearly understood my goals, I let him loose on Elance. He reviewed the full set of potential projects on the marketplace and sent my targeted proposals to each project that fit my criteria. Within a week I was getting responses back from clients and within two weeks I had been hired for multiple projects. Three months later, I had to hire my own developers because I was getting so many clients hiring me to build applications.
## The result
So how did my strategy for getting new clients as a freelancer work out? Well, here is a screenshot of my FreshBooks dashboard. Last year, my freelance business generated over $290,000 in revenue:
One month hit over $40,000:
I'm not going to pretend that this was easy. This is pretty much the opposite of a get-rich-quick kind of scheme. However, by implementing this strategy, I could cost-efficiently outsource the marketing for my freelance business so that I could focus on actual development.
# Summary
I hope that this has been a helpful guide for getting new clients as a freelancer and has given you some ideas on how you can build your own business.
# Chapter 37. Managing Client Conflicts as a Freelancer
In this chapter, I'm going to discuss _managing client conflicts_. It's simply a matter of life and business that you will run into conflicts with clients. Some of the most common confrontations seem to be:
* Going over the time you originally estimated
* Going over the budget for the project
* Not delivering a feature that matched the client's expectations
* A bug occurring in an application
None of these are fun to work through, and most of them can be avoided if the proper care is taken at each stage of the project management process. However, I want to discuss what happens when conflicts occur and how to best manage them.
# Strategies for managing client conflicts
First and foremost, do not let emotions take over. This is probably one of the hardest ones for me because I love what I do and when bad things happen in a project my first response is to get defensive, which is one of the worst responses to have. So, when I see an angry message come through from a client or take a phone call, I make sure that before I respond I sit back and try to look at the situation from the client's perspective.
It's pretty rare that a client will get upset for no reason. The majority of people are rational and prefer to stay away from confrontation, so if the client is mad there is probably a legitimate reason for it. Therefore, the first step I take is pretending that I'm the client and then I feel like I can give a better response from that perspective.
If a project went over budget and I'm imagining that I'm the client, I can understand why they're not happy because I don't like spending more money on something than I originally was told it would cost. So instead of responding with some defensive messages, such as:
_"It wasn't my fault, you were the ones that changed the scope"_
Or
_"I can't control that the fact that the designer took twice as long to deliver the mocks"_
I'll start with saying things that show that I understand their perspective, for example I'll say:
_"I completely understand how frustrating it is, the project scope has grown and it's been hard to meet all of the requirements based on the original timeline"_
Or
_"I am sorry, I should have allotted more time for the design phase, it's my fault and I will work to get it taken care of"_
If you pretend that you're a client hearing those four responses, which ones would put you more at ease? The ones where I was defensive and tried to blame everyone else or the ones where I took responsibility for the project and gave a calm reply?
The easy thing to do in a confrontation is to become defensive or respond aggressively; however, neither of those approaches will fix the issue and they'll most likely make things worse.
_Your first goal when a conflict arises between yourself and a client should be to see if you can see the issue from their perspective._
With that being said, there will be times where the core problem is the client's fault. They may be very bad at stating project requirements or they may simply be poor communicators.I once had a client who hired me a few years ago and asked me to build an application and simply gave me about 4-5 screenshots from another website and then they completely disappeared. I did my best to build what I guessed they wanted and I sent daily updates to them, and then a month later he messaged me furious that the app wasn't what he wanted and then went on to list all of the features it was missing, even though they were features he had never asked for originally.
I calmly fired him as a client on the spot and informed him that I wouldn't be able to work on the project any longer. I didn't raise my voice, I didn't explain how he gave me little to no guidance for building the app, but I knew it wasn't the type of client I could work effectively for.
I hope that this has been a good set of tips for how to manage conflict with your clients. I did quite a bit of research on this guide prior to writing it to see if there were any things that I was missing and I discovered a full library could be made out of the information based on conflict resolution.
There are discussions about personality types and more acronyms than I care to list here, however, what I've written for you in this guide is what I've used over the years and it's worked very well for me through a number of challenging situations and many different client personality types. I'm confident it will also work well for you and your clients.
# Chapter 38. Examples of Freelance Portfolios That Help Acquire New Clients
In this chapter, I'm going to discuss _examples of freelance portfolios_ that you can use to attract clients. Before deciding on the types of projects you want to include in your portfolio, it's important to answer a few key questions:
* Who will I be showing this portfolio to?
* What type of features do I love developing?
The first question is pretty standard; you need to know you your audience to ensure that your work will have its greatest impact. For example, if you want to attract small mom and pop businesses, it wouldn't make much sense to fill your portfolio with 3D Unity zombie game renderings.
The second question speaks to your passions as a developer. Too many coders create a portfolio full of generic projects that they don't truly love and it's apparent to potential employers and clients. Make sure that the projects you build fit your personality as a developer and that you are happy with the end result. Portfolio projects are not an item meant to be simply crossed off your developer checklist, they should be projects that you're personally proud of and enjoyed building.
# Examples of freelance portfolios
The following examples of freelance portfolio projects entail a comprehensive feature set and should impress a wide variety of clients. However, they are simply starting points, not hard and fast rules.
## Social network utility
This is where you build a basic social network with a clean design and features such as having posts, followers, and comments, and integrate at least one unique/advanced feature, such as giving users the ability to edit each other's posts.
When I'm looking to hire a new developer, I like to see that they know how to work comfortably with complex data models such as the ones required by a social networking application, and being able to work with advanced permissions structures is very important, so this makes a good portfolio project.
## An API tool
An example would be to develop a search engine for Stack Overflow that enables more advanced features than the main web application. A project like this would show that you can work with APIs and can implement a search engine feature.
## An accounting application
You don't have to rebuild QuickBooks. However, an accounting project can illustrate that you know how to work with financial calculations, callbacks, advanced database queries, and information security.
## A scheduling application
I've built several scheduling applications. This type of app will show that you know how to work with dates and times (which is no easy task for any developer), along with complex validations.
## A frontend application
Create an app using a frontend framework such as AngularJS or React and integrate it with a server-side backend such as Ruby on Rails. This will show that you know how to work with service-based architecture and design, which is a prerequisite for my clients.
If you build these apps (or apps that contain the same level of sophistication), you will be able to clearly showcase your expertise to potential clients, colleagues, and future employers. These projects will also give you a great code library that you can reference for future projects.
I've lost count of how many times I've reviewed past portfolio projects to see how I implemented a specific feature so I could use it on an app I was working on at the moment. I hope this list has inspired you to build out your own portfolio of projects.
# Chapter 39. Importance of Test-Driven Development for Coders
Let`s discuss the importance of **test-driven development**. First and foremost, if the terms TDD or BDD, which are short for test- and behavior-driven development, are foreign to you, they are the practice of building code tests for applications.
And even more specifically, TDD and BDD are _software development processes in which you create a test that sets an expectation before implementing any code_.
An example of using TDD to create a feature for returning a full name from a user class would be to:
1. Create a test that calls a new method, such as `full_name`, that combines the first and last name of a user and returns a string combining the names into a single value:
2. Then we'd run the test, knowing that it will fail:
3. Then we would go and add a barebones implementation of the code:
This will get the test passing:
4. Then we'd go back and refactor the implementation to ensure it conforms to best practices and that the refactor doesn't break anything:
This process is known as the **Red, Green, Refactor** workflow and is a pretty standard practice across all languages and frameworks.
# Importance of test-driven development
With that dead simple explanation of how TDD works, let's discuss the importance of TDD and the best way to answer this coding interview question.
There's quite a bit of debate on the topic, with a number of prominent developers who are against the practice of TDD, with one of the main arguments being that many coders don't use it properly and are essentially just testing pre-existing functionality instead of behavior unique to the application.
However, with that being said, if you're looking to get hired as a developer there is a very good chance that you will need to be fully versed in how to work with TDD and BDD since I don't know of very many software organizations that don't require tests.
There are four main reasons why I use TDD for all the production applications that I build or manage:
1. **Regression** : If you add a new feature into your application, you need to have 100% certainty that the new code you added won't break any pre-existing functionality in the app.
For example, if I create a new method that will break if any nil values are passed to it and I call that method on legacy data that could contain some nil values, I want to know that before the new feature goes live. Without tests, I'd have to go and manually test every part of the application each time I implement a new piece of functionality. However, if I have a full test suite I can simply run the tests, and if they're all passing it is a good indicator that the new changes can be pushed live.
2. **Team management** : If you're managing a team, having a team of developers that follows TDD processes will give you a level of transparency into what they're doing. In fact, in Kent Beck's book _Extreme Programming_ , Beck says that testing is one of the biggest keys for a development process to stay on track and budget.
3. **Documentation** : When an application that was built with TDD is finished, the tests should be able to provide 100% of the documentation for the software. Certain testing frameworks, such as RSpec, even have the ability to print out tests in a documentation form that provides a full description on the app's behavior and can be understood even by non-technical individuals.
4. **Leads the development process** : One of the most important keys to writing good software is to break code into as many small, manageable chunks as possible. When you use TDD you should naturally write small methods, efficient class definitions, and you should have minimal code bloat. When you're following TDD practices the tests themselves should lead the software's development, and the end result should be a well-organized and scalable code base.
One caveat to TDD is that testing is pointless if the test suite is not well structured. If you simply create a myriad of tests that do nothing more than test the core functionality of a language or framework, your test suite isn't going to have any benefit. However, if you build an application and let the tests drive your development and code structure, you will end up with a great application that you should be proud of.
For further information, I recommend reading the full series by _Martin Fowler_ , which is seen as one of the most in-depth discussions on TDD ever produced.
# Summary
I hope that this chapter will help you answer questions relating to the importance of TDD, and good luck with the interview!
# Chapter 40. SEO Best Practices and Strategies for Freelancers
If you build applications that users access on the web, you have most likely been asked by clients to provide an SEO-friendly website to help drive new customers to their site. You may also have realized that the world of SEO is so expansive that you could spend the rest of your career studying it and you still wouldn't have it mastered.
With that being said, I've found that if I follow a set of SEO best practices, the websites I create are search engine-friendly and clients are happy with the organic traffic sent their way. Search trends seem to change on a daily basis, and I personally don't have time to keep track of each Google search algorithm change. Instead, I focus on SEO best practices and they have worked well for me and they follow search engine white hat marketing practices so that I don't risk getting penalized by Google or Bing.
# SEO best practices tutorial
First and foremost, _content is king_. Unique quality content is always going to be the most important criteria for SEO.
## Content is king
Coming from someone who has spent countless hours studying Google's search algorithm from a computer science perspective, I can say that all the algorithm is attempting to do is connect search users with the best, most relevant content that they're looking for.
All the nuances to the algorithm, such as page ranking, counting links, and so on, are all simply ways that Google is trying to automatically find the best content on the internet. With that being said, quality content by itself is not a guarantee of traffic, but without it you won't be able to gain search engine traction for very long.
One question I get asked quite often is how many words a post should be. My answer is always " _let the content determine the wordcount_." A good rule of thumb is to have around 500-1,500 words per post, but don't waste time worrying about hitting a specific word count, focus on creating high-quality content and the word count will take care of itself.
## Creating an XML sitemap
Next, make sure that your site has a XML sitemap that you supply to Google and Bing. This will make it much easier for the search engines to index your pages. The best content in the world won't be found on a search engine if Google's spider can't find it.
## Mixing text, images, and videos
Third, users, and therefore search engines, like seeing images and video on pages. A very important criteria for following SEO best practices is to have a solid mix of text, images, and videos on a page. This helps not only for standard searches, but also for Google Image search, which can be another great source of website traffic.
## Managing your site speed
Fourth, there is site speed. Kissmetrics research shows that sites with slow load times have dramatically higher page abandonment rates, and not only is this bad for business, but your ranking with Google will decrease if your site is slow.
## Site responsiveness
Fifth is the important modern SEO best practice of site responsiveness. Responsiveness is the ability of your site to dynamically adjust in size and layout based on the device viewing it. This is about making you're your site looks and works great on, say, a standard web browser, a tablet, and a smartphone. Search engines have added responsiveness as a key criteria for site rankings, since traffic coming from mobile users is now so significant.
## Backlinks
Sixth are the backlinks to your site. Now this is a dangerous one, because backlinks used to be the top criteria that determined a site's page rank. I remember around 15 years ago when I started building websites that if I could get a popular site to link to a site I published, the new site would start getting search engine traffic and would show up higher in search results within 24 hours.
However, quite a bit has changed and now backlinks aren't as important as they used to be, but they are still very helpful. I mentioned that backlinks could be dangerous because Google has gotten ridiculously good at detecting users trying to game the system. If you use black hat techniques for acquiring backlinks you'll soon find your site penalized or even completely delisted from Google entirely.
So, what I do now for backlinks is to message journalists and other bloggers using tools such as HARO and contribute quotes and perform interviews on other sites. In return, they will typically link to one of my sites.
This is a slower way of building backlinks, but in the long run it's a good strategy. Google also watches for sites that offer to charge you to guest post and link back to your site, so I'd strongly recommend to stay away from those types of service.
## Focused content
The last for this list is for your content to be focused. If you are shooting an arrow, it helps to have a target and when you're writing a blog post or publishing a page, it's important to have a keyword to target. When I'm creating content, I select a phrase and aim to have the entire content revolve around it.
Without a focus word or phrase, it will be difficult for your post to gain traction. So, make sure you always stay focused with each post that you create. There are also plenty of great tools out there for ensuring that you're following search engine best practices. Generally, I use a tool called Yoast that uses a checkbox approach to each post.
# Summary
I hope that this has been a helpful summary of SEO best practices that you can utilize in your own projects and help you drive more traffic to you or your clients' site.
# Chapter 41. Client Communication Freelancing Tips
When I was originally building up my freelance business, I heard a common complaint from clients talking about previous developers that worked on their projects: poor communication and a lack of transparency.
As a developer, I know how easy it is to fall into the trap of wanting to dive into the code and build a project. However, without proper client communication you'll run into the following issues:
* Clients will think that no work is being performed. Regardless of reality, if you don't tell a client what you did their first assumption is going to be that you didn't do anything. This can get very messy when you send your bill and the hours that you charged don't match what the client estimated based on your updates.
* You may be building a feature in a way that the client didn't expect. I've had it happen a number of times where I understood a requirement to mean one thing, but the client had a completely different expectation in mind.
While the immediate reaction to try and fix communication issues may be to be in constant communication with the client, this approach will waste your time and it will also give them the mindset that you'll always be available, which will limit your freedom, which kills one of the main reasons you became a freelancer in the first place.
# A system to maintain proper client communication
With these issues in mind, if you can perfect this part of being a freelancer, you'll see that it leads to happy customers while also being a healthy environment for you as a freelancer.
I've put together a system for client communication that is balanced, meaning that the client will feel informed about the project but it will not inhibit your personal freedom:
1. Create a project management dashboard. I'll typically use Basecamp or Trello; however, you can use anything that you prefer and is easy for the client to use.
2. At the beginning of the day, schedule when you'll work on the project. Each morning I write down on a dry erase board all the projects I'm going to work on, and I list what time slots I'm going to work on them. For example, I'm going to work on a Rails project for XYZ client from 2 PM to 3.30 PM.
3. After the schedule is set, I message the clients and I let them know when I'll be working on their project, so they can contact me during that time slot if they need something immediate. In this way. they will also know that work is going to be performed that day.
4. After each project time slot that day, I post on the project dashboard an update on the tasks that were worked on that day. Because I like being efficient with time, I'll usually copy and paste the GitHub commit comments as bullet points.
# Summary
If you follow this system, your clients will be happy because they will know what you're doing each day, and it lets you stay in control of your schedule.
# Chapter 42. Outsource Web Developers Properly with System-Based Processes
When it comes to working on client projects, I've worked on applications ranging from apps that I could build in a few days to applications that have taken over a year of development time and involved over a dozen different developers. So, what the best way to work with outsource web developers?
Whenever I have a large project that requires a development team that I need to bring onboard, it presents a series of challenges, such as:
* Do the programmers specialize in the features that need to be built out?
* Will bringing on other developers allow the project to stay on budget?
* How can I make sure that the code quality meets the client expectations?
I could pretend that the outsource web developers I've worked with were managed properly for every project, but that would be not be telling you the truth. In fact, I decided to write this chapter based on the many times that I've had poor experiences managing development teams.
# A system to manage outsourced web developers
Based on my mixed experiences in managing outsourced web developers, I've built the following system to ensure that I have picked the right set of developers and that they are producing code that will help make clients happy:
1. **Automated testing** : Whether you work by yourself or with outsourced developers, it's vital that you use automated testing. This can include BDD or unit testing. However, this process will help to ensure that all the features of an application are working and also that new features do not break pre-existing functionality.
2. **Daily reports** : Depending on the situation that you have with your clients, it's important that you receive daily reports on the development work performed for that day. If you're not getting daily updates there is a good chance that no work was performed.
3. **Access to applications** : If you happen to only be managing the application and not actually developing, make sure that you have proper access. This includes command-line database access, your public keys on the server that the application is being deployed on, and any error logging system, such as Honeybadger or AppSignal.
# Summary
If you're bringing on an outsourced development team, it typically means that your time is limited. However, if you follow these three steps, it should help your project to be successful.
# Chapter 43. How to Create Accurate Freelance Bids?
Being able to give good estimates is one of the most critical tasks you can do as a freelancer. If you quote too low, you'll end up with an angry customer because even if you did great work, he's having to pay more than he budgeted. And if you bid too high, there's a good chance you won't get the job and the potential client will go to a competitor who gave a lower bid.
Over the years, I've been guilty of erring on both sides of the spectrum and I've had to deal with the consequences. Those consequences were not fun, which is why I've put together a formula for you in this chapter that I follow for building bids:
1. Get a detailed drill down of the project requirements. Without this you won't be able to create an accurate bid no matter what else you do. Imagine if an architect had to give an estimate to a client who said, "I want a nice house with a cool fence." He wouldn't be able to accurately estimate how much the house would cost to build. An architect will get a detailed breakdown of square footage, number of bedrooms, along with a list of all the bells and whistles. In the same way, you need a detailed breakdown of every feature that the application needs to have.
2. After you have the list of features, break them into categories, such as: database setup, frontend design, user permission configurations, and so on. And then put each of the features into one of the category buckets.
3. Put all the data into a spreadsheet segmented by the categories you created in step two.
4. Give a conservative estimate on each feature in the spreadsheet.
5. Have the spreadsheet tally up the total hours or cost and that is the project estimate.
Does this formula seem like common sense? Good, because _creating accurate freelance bids_ should be a simple process!
If you came up and asked me how long it would take to build a payroll system, I wouldn't have a clue what the bid should be. However, I do know how long building a user database will take, and I know how long it will take to implement the design, and so on. And by breaking down the project into small, specific chunks, I'm now able to feel more confident about how long the individual features will take to build.
I also update the spreadsheet as I progress through the project. That way, I'll actually have a guide for future projects. For example, if I originally estimated that building a video upload feature would take 8 hours but it ended up taking 14 hours, I will be able to more accurately estimate that feature in future projects.
# Summary
I hope that this has been a helpful chapter for learning how to systematize your process for creating estimates as a freelancer.
# Chapter 44. Freelancer Tips – Three Ways to Get New Clients
In this chapter, I'm going to discuss some strategies that I've used successfully over the years for getting new clients. There have been three main ways that I've gotten new clients:
* Outsourcing services such as oDesk and Elance (which have now merged to be Upwork)
* LinkedIn
* Referrals
So, I'll now share some of the ways that I've had success in each one of those channels so that you can take on some of the experiences I've had and apply them to your own freelance career.
# Freelancing services
Freelance services such as Upwork are great for finding new clients. The keys to success are to constantly send out proposals, dozens per day, and be quick to communicate with potential clients.
If you're a US- or UK-based developer, make sure you understand that you'll be competing from developers from all over the world, which means that the bids that you'll be competing with could be dramatically lower than what you'd want to charge. However, I've been able to get dozens of long-term clients from these services, including large organizations such as Eventbrite and Quip.
# LinkedIn
LinkedIn is an interesting tool for freelancers. I've gotten several clients from the service simply by having a filled-out profile and joining user groups for the languages and frameworks that I specialize in.
I've gotten clients such as AppDev and the Flatiron School from LinkedIn. Interestingly enough, I was never proactive with reaching out to clients. If you have a good profile showcasing your skills and you have joined enough groups, jobs start to come in. It's incremental growth, but I've found some great clients though LinkedIn.
# Referrals
Referrals are one of the best ways to get clients. This marketing channel typically takes the most time depending on your own social and client network. When I say referral, I'm not simply referencing referrals from other clients, I've gotten referrals from coworkers, friends, family, and through networking events, such as local Chamber of Commerce organizations.
# Summary
I hope this has been a helpful chapter to different freelancer strategies to grow your client base. If you're an employer who works with freelancers, then I hope you also read this section closely, and it gives you a better insight into freelancers' minds, and how to get the most from them.
# Part III. Career Skills
# Chapter 45. Should I Learn to Code? – A Balanced Perspective on Programming
To start off this part of the book that covers all about what it takes to have a successful career as a developer, I'm going to talk about the question: should I learn to code?
This question was sparked by the controversial post on TechCrunch by Basel Farag (<https://techcrunch.com/2016/05/10/please-dont-learn-to-code/>), where he urges people to not learn coding. As the founder of Rails bootcamp you may think that my reaction would be to spout off a full list of all the reasons why everyone should learn how to code, because that's good for business, right?
However, I went through his post in detail and I also reviewed a slew of response posts that called for Basel's head, denouncing him as a modern-day heretic.
# Should I learn to code? – a balanced look at both sides
After going through all the content around Basel Farag's argument, my belief is that both sides of the argument have valid points that should be considered by anyone deciding whether to learn to code or not!
Let's first explore the idea that _not everyone_ should learn to program. Farag proposes that coding is hard, which it is, and therefore, the dream that you can take a few online tutorials and become a professional developer is a lie... and he's 100% right about that. Development, especially for true production applications, is very hard and takes years of study to become truly proficient. It's not enough to simply learn how to build an application that lets users create records in a database from a form, for example. A professional developer needs to have expertise in:
* Managing dependencies between code libraries
* Working with object-oriented programming best practices
* Having clean code
* Implementing automated testing for continuous development
* Knowing how to transition seamlessly between various frameworks
And the list goes on and on...
In fact, even though I'm a professional developer and have worked for clients as big as Eventbrite and Chevron, I spend several hours a day going through development books and online guides to simply keep up with all of the new coding techniques and systems that are continually emerging.
So, all that is to say that Basel Farag is right in that if you want to become a professional developer, you have a challenging road ahead of you. Of course, if you make it through, you will have gained a tremendous skill and you could well find yourself in demand by a number of industries.
Let's now explore the opposite side of the argument, that everyone should learn how to code. I really appreciate the VentureBeat article by Edward Chiu (<https://venturebeat.com/2016/05/22/how-coding-kick-started-my-sales-career/>), where he describes how he went through a coding bootcamp and he did not become a professional developer... but that he did get a great job as a sales engineer that he never would have gotten if he wouldn't have learned coding!
This is the side of the argument that is ignored all too often, because learning to code does not mean that you have to become a professional developer, but it will give you a new skill set that can be used across many different jobs. Steve Jobs said it best when he said:
> "I think everybody in this country should learn how to program a computer because it teaches you how to think."
To me, Jobs hit the nail on the head here: learning how to code is not simply a prerequisite for becoming a professional developer, even though that can happen as well. It gives your mind a system for structured thinking that you can apply across all disciplines.
From a personal example, I used to have a hard time taking notes. I would randomly write down words all over a page without any really coherent flow and when I'd look back at them later they would be completely worthless to study from. However, after I learned to code I started actually taking my notes in code form to give them structure, I'd set up loops for repeating items, classes to hold a topic's attributes and processes, and return statements for the end results of whatever I was trying to learn. That may sound nerdy, and it probably is, but it gave me an organized system for taking notes and learning new topics.
# Summary
In summary, I hope that this balanced view has help you answer the question: should I learn to code?
# Chapter 46. Following Your Passion – Good or Bad Advice for Developers?
As a developer, should you follow your passion? I hear this term and this advice quite often and I'm not a fan of taking things at face value, so I wanted to look into the concept and review it here.
At a high level, the advice centers around the idea that if you do work in a field that you truly love, you will be happy. Many people point to Steve Job's Stanford commencement speech when he says:
> " And most important, have the courage to follow your heart and intuition. They somehow already know what you truly want to become."
# Following your passion – a case study
While writing and researching this book, I came across Cal Newport's book, _So Good They Can't Ignore You_ , where he took a deep look at Steve Job's life. He shows that if Jobs would have spent his life adhering to the advice of following your passion, he never would have started Apple, but instead would have become a Buddhist monk living in Asia.
Jobs started Apple for the practical reason that he saw a way to make some quick money selling a set of computers to a local electronics store, and that sale sparked his interest in technology. So, it seems like following your passions is not a cut and dried process of doing work that you love.
However, my advice to you is not to ignore your passions, but instead to use them to build a career that you truly love and can excel in.
Sticking with the Steve Jobs example, one of the other passions Jobs had was calligraphy, which is essentially fancy handwriting if you've never seen it before. When creating the user interface for Apple's operating system, Jobs leveraged his extensive expertise in calligraphy to integrate fonts into the programs. Here was an example of Jobs combining his passions with a practical implementation.
As developers, we're in a unique position where we can work in a wide assortment of industries: if you love sports you can work for a professional sports team, or if entertainment is your flavor then you can work for a media company. The possibilities are really endless.
One of my biggest passions is baseball. I grew up around the game and I love everything about it. I wasn't skilled enough to play professionally, but right now I'm working towards my PhD in computer science and my topic of research is big data analysis in baseball.
It wouldn't have been practical for me to waste who knows how many years trying to make it as a professional baseball player, but I can leverage my skill as a developer and still perform work that I love to do. I've also have the privilege of having multiple clients that are in the sports industry, which has allowed me to work in the sector that I'm passionate about while still leveraging the skills I have as a developer.
# Summary
Hopefully, my research and personal experiences can help you decide if following your passion is the right decision, or if there is a middle group that would lead to a better end result.
# Chapter 47. How to Learn to Code from Scratch? – A Practical Strategy
Becoming a developer is a rewarding yet challenging task. One of the greatest blocks for people to understand programming is simply having a plan and deciding where to start.
In this chapter, I'm doing to walk through strategies to help you learn how to code from scratch. I've been a developer for a number of years. I taught myself how to code and I've witnessed a wide variety of educational techniques for learning programming over the past decade.
Some of the strategies I've seen are good, others are a waste of time. This chapter contains the strategies that have stood the test of time and will help you launch your coding journey.
# Small bites
First and foremost on the list of tips to learn how to code from scratch is the principle of **small bites**.
I have a friend who trains professional and Olympic athletes for Adidas, named Mark Verstegen. Back when I used to train at his institute, he would always say something that really stuck with me. When any athlete presented a tough goal, such as qualifying for the Olympics or making it to the big leagues, he'd ask them:
> "How would you eat an elephant?"
After the athlete would look at him with a confused look, he'd follow by saying:
> "It's not a trick question, the only way to eat an elephant is one bite at a time."
This is great advice for many aspects of life. However, I've discovered that it's an especially important concept for developers to understand. When I think back to when I was learning development, my greatest obstacles and challenges came when I tried to do too much.
For example, when I was trying to build a new feature I would attempt to code the entire feature at once. Most of the time this would end up with the program not working, and then I'd have to go through every line of code until I figured out what was wrong.
However, the more experienced I've become as a developer the more I realize the importance of breaking concepts down into small, easy-to-manage chunks.
Let's imagine that you are building a connection to the Twitter API. Instead of trying to build the entire feature, focus first on connecting to the API. Then print the values returned from Twitter. Finally, you can format the data so that it looks nice. By breaking what you're learning into small components, you'll discover that you will have a better understanding of the processes going on. You will also be able to remember how to implement the features later in real-world projects because the concepts will be more tangible.
# Tutorials
Over the past few years the online educational space has grown exponentially. Whether you are looking to learn Java or Ruby, you'll be able to find countless tutorials that will help you understand programming. These types of tools most likely won't turn you into a professional developer by themselves, since achieving a professional level of skill takes years and typically requires you to work on a wide range of real-world projects.
However, tutorials can be a great introduction to programming. In addition to giving step-by-step guides for how to build applications, screencasts are also great for showing you what types of apps a specific language or framework can build. When I'm learning a new language I'll watch a full series of tutorials without even trying to type in the code. I do this so that I can familiarize myself with the capabilities of the language.
One of the weaknesses with tutorials is that it's hard for them to replicate your own environment. For example, if you're working on a Java programming language tutorial from a few years ago, there's a good chance that the instructor will have a different language version than you do. This will cause some confusing bugs, and without any assistance many individuals have quit their programming dreams out of frustration.
But don't let that scare you away from using tutorials. I credit a number of tutorials with helping me teach myself development. And I highly recommend them as a great place to start, especially when you want to learn how to code from scratch.
# Reading
Next on the list is reading. Libraries could be filled to the brim with the number of programming books that are on the market. I have even written a few!
I like going through coding books because they allow me to go at my own pace. When I go through video tutorials, it usually means that I need to dedicate a specific amount of time to go through the videos each day. However, with a book I can read a few paragraphs or I can go through a few chapters. When you have a full-time job and you're learning programming on the side books are a great resource.
This is because they allow you to learn at your own pace. Books can also be a good resource later when you need to reference a specific topic. Also, when you go through a programming book I highly recommend you write and run the code from the book. This will help you remember the programming language syntax much better than simply reading it.
Remember that reading retention is incredibly low in most individuals. However, if you combine reading with actually writing the code as you're going through the content, you'll see much better results.
Another trick to use when reading programming books is to not look at the book when you're writing the code. For example, if you are reading my Ruby programming book you'll see a code snippet when you're learning how to use object-oriented programming. If you force yourself to type the code without looking at the book the entire time, you'll discover that your retention will increase dramatically.
# Real-world projects
Last on the list to learn how to code from scratch is building real-world projects. After you've gone through a number of tutorials and read a few books, you'll be ready to try your hand at building applications.
A natural question to ask is: _"What types of projects should I build?"_ There's really no right or wrong answer to this question. If you have an idea for a business then you could start with trying to build it with your newfound coding knowledge.
You could also look at re-building current applications, such as creating a Pinterest clone. I've found this technique of creating cloned sites very beneficial since it allowed me to focus on building functionality instead of having to waste time on coming up with ideas.
For example, when I learned the Swift programming language, I built an Instagram clone. Years ago, when I was learning HTML and CSS, I recreated the Google homepage from scratch. The most important factor to remember about building real-world projects is to stretch yourself. No developer ever improved by duplicating functionality they are already comfortable building. Instead, make sure you are challenging yourself to implement features that you've never created before.
# Coding is hard
On a final note, don't let anyone tell you different: coding is hard! From setting up a development environment to building functional applications, programming will greet you with challenges at every stage.
# But you can learn programming
However, with that being said, you can become a developer. There's not a magical programmer gene that coders are born with. It simply comes down to:
* How determined you are
* If you're willing to work consistently
* How good your strategy is when it comes to learning
# Chapter 48. How to Choose a Developer Specialty?
Through many years of training developers, I've discovered that it's vital for coders to decide about their specialty and focus. In this chapter, I'll share with you how I personally view the different developer directions and areas of expertise that are available.
You'll quite possibly make this choice more than once in your career as a developer, so I recommend that you always observe and be aware of the choices that can be made in your career.
# How to choose a developer specialty?
The world of software development is so vast that it's impossible for someone to master every aspect of the process. Consider if I approached a world-class track and field coach and said that I wanted to train for the Olympics and win a gold medal in track and field. The coach would most likely give me a once over and chuckle to himself. But after that his first question would be: _"What event do you want to train in?"_ He would ask this question because the training regime for the 100-yard dash is dramatically different from the high jump.
In the same way, as a developer, you need at least once, and quite possibly several times, to narrow your focus on what type of developer you want to be in the years ahead.
I'm going to walk you through each type of developer category so you can see what they entail.
## #1 – the full stack developer
In deciding how to choose a developer specialty I always like to start off with the full stack option. I start with this option because many new coding students I've spoken with assume that all developers are full stack developers. And this is simply not true.
Full stack development means that you feel comfortable working with every stage of an application's development. Referencing our track and field analogy, a full stack developer would be like a decathlete. This is the category that I personally fall into. My focus on the full stack side of programming is due to a number of factors:
* To teach students and write development curriculum I need to be familiar with all of the key development types.
* I've spent years as a freelance developer. And in many cases, freelance coders are asked to build an application from the ground up, create all the features, design the system, and deploy it to the web or app store.
Full stack developers need to be a jack of all trades! Much like a decathlete, full stack developers are usually good at a number of technologies. However, a common pattern you'll see is that it's very difficult to be world class at _every_ layer of the development stack.
Programming is simply too complex, and languages/frameworks change versions so rapidly that it makes it nearly impossible to excel at every stage of the app development life cycle. Because of how time consuming each level of the development process is, full stack developers simply don't have the time to become true masters at any one aspect.
As a full stack developer myself, I mitigate this issue by focusing my time on the components that I excel in, such as server-side development, and then working with other developers to help cover my weaker areas, such as UI/UX.
## #2 – the server-side developer
Next on the list of developer types is server-side programming. This is probably my favorite layer of the developer stack. Server-side specialists spend most of their time working on building and implementing algorithms that enable programs to work properly.
Additionally, server-side developers typically spend quite a bit of time building APIs. This is because most server-based applications need to communicate with the outside world in some form or another. This layer of the development stack will require you to specialize in a language, such as Ruby, Python, Java, or C++.
## #3 – the frontend developer
When it comes to choosing a developer specialty, the third layer to choose from is the frontend component. Not too long ago a frontend developer was considered someone who spent all day working with HTML and CSS. Their main goal was to make applications look _pretty_.
However, the definition of a frontend developer has changed dramatically with the advent of client-side frameworks. These frameworks, such as Angular and React, have made it possible for frontend programmers to build complete apps with little server-side interaction.
These applications are rendered completely in the browser because they're written in JavaScript. And whenever the app needs to get additional data it simply communicates with APIs. A common pattern that I work with is building a number of server-side Ruby applications and then having a single Angular frontend app that renders the user interface in the browser.
So, if you love building applications that users will directly interact with and the idea of working with APIs doesn't scare you off, frontend development might be the right choice for you.
## #4 – the mobile developer
Next on the list of developer types is mobile. If the idea of building the next Angry Birds or Instagram excites you, the mobile development field may be a good fit.
Mobile programming used to be a very difficult field to enter. Only a few years ago you would have had to master multiple languages (Objective C and Java) to build smartphone apps. However, JavaScript frameworks such as Ionic and React Native have made it possible to use JavaScript to build apps that behave like native smartphone applications.
You can still use languages such as Swift, Objective C, and Java to build truly native applications. And there will always be a great set of jobs for developers who specialize in these languages. However, if you are a freelance or full stack developer, by leveraging a JavaScript framework you can build smartphone and tablet-based apps for all platforms.
And it's been my experience that the learning curve for these JavaScript frameworks is quite a bit lower than the traditional mobile languages. Additionally, you may have noticed that the tools used for JavaScript-based mobile apps and frontend programming are similar. Because of this synergy, I have had a number of developer friends who have moved away from server-side development and moved into frontend coding because it allows them to tackle building applications for desktops, tablets, and phones.
## #5 – the data scientist
This used to be considered the data field, and a few years ago I'd have called this something like the data developer category. However, data and big data have rapidly morphed into the fields of statistic data analysis and using artificial intelligence, such as neural networks, to gain insight into the huge amount of information now available.
These new fields are changing the face of how we process data and understand information, and it's a huge new career field for developers to explore. Right now, the mathematics involved can be quite intimidating, so you'll need to decide if you're ready for some heavy math and deep algorithmic learning. The math is going to become more abstracted over time, and if you're drawn to the idea of artificial intelligence and deep algorithmic learning systems, this is certain an area to consider for an ambitious developer today. Some of the most modern Python libraries provide an excellent way for a developer to immerse themselves into this field.
# Making the decision
If you are new to development, don't feel pressured to pick out a specialty immediately. Instead, my recommendation is to explore each type of development layer until you find a focus that you truly love.
In this chapter, I've provided a very high-level view of the developer types. However, in reality, you will need to become even more specific with your development focus.
For example, if you're a server-side developer, you may want to focus on building eCommerce applications or implementing accounting systems. If you are an aspiring frontend developer, you may want to become a world-class security specialist.
A key insight that I've discovered helps quite a few people, especially newer programmers, is to look at developer job boards. Job boards are great for listing out the specialties that companies are hiring for. And by going through a list of potential job descriptions it may help you figure out what you want to focus on next in your career.
# Chapter 49. How to Choose Your Next Programming Language?
We have discussed the importance of picking a development specialty, such as frontend, server side, or mobile. However, simply choosing a specialty is not enough. You also need to decide on what programming language you want to focus on for the initial, or next, stage of your career.
No book or guide can tell you what language you should learn next, of course. That's a decision that can be made only by your. So, in this chapter my goal is to help you decide how to pick a programming language based on your current objectives.
# How to pick a programming language?
If you went through the exercise for picking your developer specialty, you will notice a similar pattern for deciding on a programming language. A quick perusal of Wikipedia will reveal that there are literally hundreds of programming languages to choose from. If you take the approach of looking at each language one at a time, you might be able to make an informed decision sometime in the next hundred or so years!
Since iteratively going through the full list of languages isn't practical, I recommend two processes for helping you decide how to pick a programming language.
## The next job you want
First on the list is basing your programming language choice on the job you want. If you've heard the phrase _"dress for the job you want"_ , it also applies to development.
For example, if you want to work for Microsoft or with Microsoft-based products it wouldn't make sense for you to spends years learning Python and Django. Instead, you will want to focus on learning the .NET development stack and languages such as C#.
This strategy can be smart in certain cases, such as with Microsoft, since there are countless .NET framework positions available on the job market. This means that even if you can't get hired on with Microsoft, you can still get a job for an organization that utilizes the .NET stack.
However, this approach doesn't work quite as well for more specialized languages and companies. For example, imagine that you spend years learning Facebook's flavor of PHP, codenamed HipHop. If you fail to get a job working for Facebook you will discover that not very many companies utilize the HipHop framework and your job opportunities will be limited. I view this approach as a bit risky because it tends to place all your eggs in one basket.
## Your specialty
My personal favorite approach is to pick a language that fits in with your development specialty. I'm partial to this strategy because it's what I used in my development journey and it worked quite well for me.
Deciding how to pick a programming language based on your development specialty means that you look at the types of applications you want to build. And then you work backwards to put together a list of languages that are best suited for your objectives.
For example, let's take the case study of you deciding that your development specialty is going to be building big data applications. By taking this approach, you can dramatically narrow down the list of programming languages that fit with your goals. Many languages can perform big data processing, but only a few languages truly specialize in it, such as:
* R
* Scala
* Python
So, by looking at your specialty first, you have just narrowed down the list of languages from thousands... down to three.
### Specialty-based mapping
Since I find this approach to be the most effective, let's walk through a mapping of development specialties to popular languages:
* **Full stack development** : If you want to be a full stack developer, the Ruby on Rails stack may be the best choice for you. This stack offers a great set of tools for web developers and allows for the build out of robust applications.
* **Frontend development** : For frontend developers the path to follow resides on the JavaScript track. The JavaScript programming language has emerged as the clear winner in the frontend development space. And by becoming fluent in JavaScript you will be able to work with popular frontend frameworks such as React and Angular.
* **Server-side development** : The world of server-side development can be a bit intimidating. If you review the server-side languages you'll discover lower-level languages such as C, C++, and Java. However, programming languages such as Ruby, Python, and Go also specialize in server-side development.
* **The data scientist** : The world of data science can appear at first sight to have impossible demands on a developer to also be a high mathematician. However, if you explore the modern Python libraries, you'll find a rich set of ready-made algorithms so you can be creating your own neural networks and machine learning systems very quickly indeed. I'd recommend Python for this reason if you're already familiar with the language. If you have a bit more of a statistical background, then I'd recommend that you explore further how the R language can get a gateway into a new career path.
Through my programming journey, I have worked to specialize in one interpreted language and one compiled language. I chose Ruby for my interpreted language, and for my compiled language I started with C. However, I haven't found very many practical uses for my C knowledge over the years. Also, I have moved onto the Scala language since it is a good fit for building big data algorithms.
# Summary
In summary, as with picking your development specialty, my recommendation to aspiring students is to experiment with a number of languages before deciding on which ones to learn extensively. There are a number of tools online that make it helpful to see a side by side comparison for various languages.
For example, one of the deciding factors that led me to learning Scala was looking through the Scala algorithm implementations on Rosetta code (<https://rosettacode.org/wiki/Category:Scala>). So, don't rush into picking out a programming language. Give the decision plenty of thought and research and you will put yourself in a better position for making the right choice.
# Chapter 50. Developer Soft Skills – Learning How to Gain an Edge in the Marketplace
I've talked quite a bit in this book about improving as a developer. Most of the time I focus on how you can learn new technical skills, such as becoming more proficient in a programming language or framework.
However, if you limit your knowledge to technical talent you will be decreasing your chances for success in the marketplace. In this chapter, I'm going to walk through five key developer soft skills that you can utilize to become a well-rounded coder.
# Developer soft skills
The list of developer soft skills I'm about to present you is by no means comprehensive. But what it does offer you is a representation of the soft skills that I've personally used and had success with.
From being the IT Director of a national energy company in my late 20s to the CTO of a coding bootcamp with locations around the world, I've seen these skills help me at every level of my career. And as you'll notice, they have very little to do with actual technical ability.
As a caveat, I do not mean for this list to overshadow skills such as practicing clean coding habits or focusing on improving as a developer. Instead these skills should complement your engineering talent.
## Writing
First on the list is the ability to write. In the book _Rework_ , by _Jason Fried and David Hansson_ , who are also the founders of Basecamp and the Ruby on Rails framework, wrote that one of the skills they look for in job candidates is their ability to write. This includes positions that you would think writing skill would be pointless, such as developers and system administrators.
Obviously, a developer needs to be able to be skilled as a coder. However, the book explains that if a developer can write, it is a sign that he or she is a good communicator. Writing skill doesn't mean that each memo you write has to sound like a riveting novel. Instead it means that:
* You can organize your thoughts properly
* You can communicate what you want to say so that others can understand
## Conversation
Next on the list of developer soft skills is the ability to converse well with others. Now if you're like me, this is by far the most challenging skill on this list. If I had my way I'd stay behind my desk building applications all day and never interact with another human. It's simply the way I was wired, and I know I'm not alone in that desire.
However, conversational skills are an absolute requirement when it comes to advancing in your career. Whether you are a freelance developer looking for new clients or a software engineer looking to get promoted, you'll discover that the top prerequisite to your success is not technical skill, it's likability.
If someone likes you they are going to want to give you a chance to succeed. And one of the most straightforward ways to get people to like you is by becoming a good conversationalist.
### Conversation tips
Thankfully, I've discovered that the system for having great conversations is pretty straightforward. Here are some tips that I've used to improve at this skill:
* _Think back through your life and come up with some entertaining stories about yourself_. People love stories, especially if they are funny. And I've discovered that telling a few well-timed stories has been able to get me in the good graces with CEOs and executives over the years.
* _Make the focus of conversation be on the other person_. People love talking about themselves, so by simply asking insightful questions you will be considered a great conversationalist... even though you let the other person do all the talking! Like your stories, come up with a list of questions that you can recall at a moment's notice.
* _Do not complain_. I have yet to find the person that likes to converse with someone who constantly complains. With that being said, countless people seem to enjoy bringing up every negative thing that has happened in their lives when they meet someone. The good thing about this is that if you can have a conversation without complaining you will stand out as being an upbeat and likable person.
## Management
Moving down the list of developer soft skills, the next item is management. Now if you're an entry-level developer, don't tune out. Management doesn't have to mean managing people or projects.
When I say management, I'm referring to how you attack each task you're given. For example, if you are handed a new feature to build, do you jump right in and start coding? Or are you more organized with your approach? If a client or managers see that you take a systematic approach to every task you're given they are going to feel more confident giving you more responsibility.
To improve this skill, I recommend you read up on project management books or take an online course on the topic. The few days that you'll spend learning about management practices will help serve you well the rest of your career.
## Design
Next on the list is design. In Scott Adams's (the creator of Dilbert) book _How to Fail at Almost Everything and Still Win Big_ , Adams describes how knowing the basic fundamentals of design should be required knowledge for all engineers.
I can't tell you how many times I've heard a developer say something like _"design really isn't my thing"_. It's fine if your top skill isn't design. However, learning the basics of what qualifies as a well-crafted design takes such little work that anyone who doesn't learn about it is simply being lazy.
As a developer, if you haven't researched what it takes to create a good design, you are going to be quickly bypassed by others who read a single book on the topic. There have been multiple times early on in my career where I neglected design and it cost me dearly.
I remember one time where I spent weeks building an incredibly complex feature to only have management spend the entire meeting talk about how much they hated the design, while completely ignoring the actual functionality. If I would have spent a few hours to design the look and feel of the product the meeting and project would have had a more favorable result.
## Public speaking
Last on the list of developer soft skills is public speaking. This may seem like a useless skill for a software developer. However, let me give you two scenarios to think about:
* In scenario 1 there is a brilliant developer with poor public speaking skills. When asked to present a project that he built, an incredibly well-built project I may add, the developer talks in a monotone voice during the whole presentation and the product demo is filled with him simply moving from one page to another.
* In scenario 2 there is another great developer. But this engineer has worked on his public speaking skills and gives a well-organized demo. His time in front of the room is filled with clear language, amusing anecdotes, and analogies for each feature to make the project understandable for everyone in the room.
If you were in the room, which one of the projects would seem more appealing? It doesn't take a MBA to know that the developer in scenario #2 will win each time.
Note that both the projects were great pieces of software. Like I mentioned earlier, soft skills are not a replacement for technical skills. They are something to layer on top of programming expertise. You could give a Steve Jobs-level speech, but if the product doesn't work it won't matter.
### Becoming a better public speaker
Public speaking is ranked as one of the most feared tasks to perform. However, I can tell you from experience that you can improve at public speaking quite easily. There are two things that I've done to become a better speaker:
* I am a member at a local Toastmasters group. Each week I attend a group meeting where I can practice getting up and talking in front of a group of people. By simply forcing myself to practice this skill consistently my public speaking ability has improved dramatically.
* Additionally, I listen to one TED talk each day. The TED conference lectures are given by some of the most skilled orators in the world. By listening to a new talk each day, it has helped give me ideas of ways that I can craft my own speeches and it has helped to build a mental model for what makes a great speech.
# The importance of soft skills
So, now that you know the list. How important are developer soft skills? I can't tell you how many times I've seen an inferior developer promoted to management simply based on their ability to speak well in meetings or converse with co-workers.
Remember that the key to each of the skills on this list is that they help people feel more comfortable being around you and that they will be confident that you can get the job done. Likability and confidence are two key prerequisites you'll need to gain an edge in the marketplace.
# Chapter 51. Developer Learning Options – A Practical Analysis
The entire world seems to be talking about the importance of becoming a coder. However, many of these discussions aren't practical. This chapter will walk through the various developer learning options available today, and help you decide on which option is right for you and your goals.
# Degrees of programming expertise
If you're new to development the first task you should complete is to decide why you want to learn programming. There are a number of reasons for learning how to code, including:
* Becoming a professional developer
* Improving your skill in your current profession
* Learning for fun or for hobby projects
It's important to decide on your goals since they will dictate what learning strategy to take when it comes to how to learn programming. Let's look at each of them.
# Becoming a professional developer
If your goal is to become a professional developer, congratulations! You'll be joining one of the fastest growing industries that the world has ever seen. The pay is great and the right developer jobs are both fun and rewarding. However, with all those benefits comes a strict set of learning requirements.
Pro coding jobs will require you to become proficient in a programming language and several frameworks. And simply building applications that function properly isn't enough for most jobs. Instead, you'll need to be able to build apps that follow processes such as:
* Test- and behavior-driven development
* Clean coding styles that are scalable and adhere to industry best practices
* Coordinate with developer teams and seamlessly work with code version control systems
And the list goes on and on...
## Developer bootcamps
Any platforms or guides that say you can become a professional developer in a month or claims like that are simply lying to you. Becoming a professional developer takes years. I've been a developer for over a decade and I'm still learning each day. With that being said, if you want to make a career as a developer there are countless resources for achieving your goal.
Personally, I'd recommend starting with a developer bootcamp. Bootcamps allow you to become fully immersed in a language or framework over a course of several months. And many of them, including the one I run, DevCamp, offer job guarantees after you've completed the course.
These immersive programs aren't easy or cheap. They'll usually require around 40 hours a week of study and practice. And you'll find they range from around $5k to $20k. This may seem like a pricey option; however, how many training platforms can guarantee that you'll get a job after a few months of study?
## Is this practical?
It may seem like I said a contradictory statement when I said that developer bootcamps can help you get a coding job in a few months. But I also said it takes years to become a professional developer. Let me let you in on a dirty little secret in the software world... many individuals working in developer jobs aren't professional programmers!
After getting hired from a coding bootcamp you'll most likely be a junior developer. The companies that hire bootcamp graduates understand that they're usually new to programming and they take it upon themselves to continue the new hire's education.
This is a win-win scenario. Developers can get great paying jobs with limited experience. And software companies can train developers to follow the procedures specific to their organization.
# Improving your skill in your current profession
Another great reason for taking interest in how to learn programming is to improve in your current career. I'll give you a great example of how this can work. I have a good friend who spent his whole education focused on business. After graduating from college, he got a job working for an energy company and he realized he wasn't climbing the corporate ladder quite as fast as he wanted. He approached me initially to see what it would take for him to transition and become a full-time developer. After a few minutes of speaking with him it became apparent that he didn't really want to become a programmer, he was simply frustrated with his current job.
So, I began asking him questions about what types of tasks he had at work. He worked in the supply chain division for his company and it was his job to comb through fleet management data and generate reports. I proposed that he learn the basics of programming and then to focus on data science.
After a few months, he had successfully built a big data analysis program that was able to manage his entire division. When he presented his work to the company's executives they were so impressed with the program that he was promoted and given a hefty raise. When I talk to him now he doesn't mention leaving his job anymore. Instead he discusses how he loves it and how he's constantly looking for new ways to integrate automated mechanisms to improve the work he does.
## Is this practical?
If you think this scenario fits with your goals. You can follow a much different path than professional developers. You can take online courses that walk you through practical projects that you can re-purpose for your own needs.
For example, the executive I just mentioned didn't learn how to build a machine learning algorithm from scratch. In fact, he'd be completely lost in even an entry level computer science class. Instead he took one of my programming courses and saw how the decision tree I built could be used in his division. After altering the data points, he could use it for his specific needs.
# Learning for fun or as a hobby
This is a fun and relaxed way to learn. If you have a pet project that you've been wanting to build or if you simply like learning for the sake of learning, you'll discover a wide variety of resources that will help you achieve your goals.
When people ask me where to start when it comes to building code projects for fun, I typically point them to practical tutorials. If you're simply learning for fun you don't have to waste your time on complex computer science topics. Instead, you can focus on following step-by-step guides that walk you through how to build projects.
For these types of guides, you can access affordable ones on sites such as Udemy. The great thing about these types of courses is that they come with:
* Videos
* Written guides
* The source code for the project that you'll be building
This approach to learning makes for a great way to be introduced to development. And you never know, starting with hobby projects could end up with you going to the next level and learning professional programming.
# Summary
In summary, I hope that this has been a helpful discussion on the various developer learning options that are available. I'd recommend checking out each option and taking an honest look at which one is right for you. No matter what you decide, learning how to code is a great experience and I recommend everyone to study it in one form or another.
# Chapter 52. Is it Possible to Lose Your Coding Skills?
This topic was inspired by a Boing Boing article (<http://boingboing.net/2016/06/08/coder-fired-after-6-years-for.html>) that tells the story of a QA developer who spent 6 years working for a company and literally did nothing besides playing computer games, browsing Reddit, and wasting his time.
The Reddit user, FiletOfFish1066, posted his story to the site and it instantly became a supreme case study for what happens when you don't continue to develop your coding skills.
After not working for 6 years, he says that he has completely forgotten how to develop and now he's out of a job. The story goes that he started working as a software tester and realized that, by writing some scripts, he could fully automate his job. So, he worked for about 8 months building testing scripts and after that he simply let the tests run. He didn't have to do anything besides kick back and play video games all day. There are a number of lessons that developers can learn from this story and that's what I want to cover.
I don't really blame the employee in this case. Apparently, there was such little oversight in his organization that he could get away with not working for 6 years. That tells me that the company he worked for has serious structural problems to let that go on for so long. It should be noted that many organizations are employing people to do work that software can do, but instead are choosing to waste money with archaic manual processes.
This developer started his job doing exactly what he should have done. He recognized that there was a way to automate his job, which was a great first step. If he would have gone to the management and shown what his automated script did, he would have most likely been promoted for his expertise and would have a great career right now.
**It's possible to lose your coding skills**. If a world class bodybuilder would stop going to the gym, eventually he'd lose his muscles. In the same way if you stop honing your craft as a developer, you'll eventually lose your programming skills.
A few years ago, I met with a gentleman who was the vice president of software development at a large energy services company. He got into the position by selling software that he had built himself, which has become the industry standard and currently processes billions of dollars in transactions each month. Even though this individual used to be at the top of his field as a developer, since he became an executive, he got further and further away from coding the application, and he admitted to me that he wouldn't even be able to build a simple program now.
I just finished reading the book _Peak_ by _Anders Ericsson_ and _Robert Pool_ , and it gives case studies from the medical field, that show that the most experienced general practice doctors are not always the best in their field compared with less experienced physicians. The book explains that, on average, the longer a general practice doctor has been working in industry, the less they focus on learning and therefore they start to lose some of their expertise.
Surprisingly, it's actually doctors who have recently finished fellowships and gone through extensive training that perform the best. The authors did note exceptions when it came to specialists. For example, cancer specialists that perform surgeries daily and are constantly working on their craft, perform better than less experienced surgeons.
I thought that was a great example for developers because I know, from my own experience, that if I'm not daily using my development skills, they will atrophy. There really is no middle ground, when it comes to development you're either getting better or worse, you won't stay the same.
So how can you ensure that you're always improving and that you won't lose your coding skills? Here are a few practical tips that I use.
**Learn something new about development each day** and be intentional with how you learn. In fact, just yesterday I taught myself how to integrate growl notifications into a Rails application (<https://rails.devcamp.com/professional-rails-development-course/advanced-user-features/how-to-integrate-growl-type-notifications-rails-app#.V2G-uD4TDUQ.twitter>), which was something I hadn't done before.
**If you simply repeat the work you've done in the past, you won't improve**. Thinking back to our case study of the general practice doctors, the research showed that the reason why the older physicians skills decreased was because they performed the same work day after day and eventually the only tasks they could perform properly were the things that they had repeated each day.
To be 100% honest, this part isn't very fun, which is why only a few people do it. Learning new and challenging skills can be intimidating and stressful. In the same book _Peak_ , the authors said that a common trait among all top performers, in every field they researched, was that they were willing to deliberately practice skills that they found difficult, because it was only by mastering those skills that they could grow in their profession.
A practical way for developers to implement this method of deliberate practice is to write down a list of features that you have never built into an application before. Then, spend time each day until you have successfully built each component. After you're done with that list of features, create another list. I personally have a list that I work and study from and it's helped me to feel confident that I'm learning something new each day and that I'm constantly improving as a coder.
On a final side note, I've had multiple people message me about the story from Boing Boing asking how it was possible that the developer created a script for automating his work. Without further details, it's impossible to know for sure; however, since he was in the testing department my guess is that he built a test suite, using tools such as Capybara (<https://rails.devcamp.com/professional-rails-development-course/application-build/bdd-index-view>) to run through the software and generate reports on features. So that would be my guess.
# Summary
I hope that this guide has helped to inspire you to be deliberate with your practice and that you will continually improve as a developer.
# Chapter 53. Is Writing Bad Code Immoral for Developers?
In this chapter I'm going to discuss a slightly odd question: is writing bad code immoral? This leads to the concept of the importance of developing well-written code.
This may seem like a weird question to ask because the mindset of most developers is that code projects are neither moral or immoral, they're simply programming files that perform various functionalities. I would like to think that most developers take pride in their work and therefore want to write code that adheres to best practices. However, given schedule and budget constraints many projects devolve, with the top goal becoming _to simply work_ and being completed as soon as humanly possible.
However, this mindset can lead to issues such as: missing edge cases for features and poorly organized codebases that are difficult to maintain. Regarding the question of _is writing bad code immoral?_ , I heard a great story from one of my Computer Science professors at Texas Tech, Dr. Michael Gelfond, which is where I got the idea to write this chapter.
During one of our lectures, Dr. Gelfond posed the question and then told a story. A few decades ago when he was a programmer working for a software organization, he ran into a nasty code bug. It took him several days to figure out that the previous developer had built a poorly constructed function that was causing the module that he was working on to break.
After Dr. Gelfond told us the story, he asked us again if writing bad code was immoral; most of the class answered that it wasn't. But then he asked if we murder someone a few days before they were going to die, "is that immoral?" To this, everyone answered with a unanimous "yes." He finished his lecture by saying, "Well wasn't it immoral that the last developer's code stole two days away from my life?"
That story and question has stuck with me for years, and now my answer to the question "is writing bad code immoral?" is a resounding "yes"! As developers, we should take pride in the work that we produce, not just for our clients' or employers' interests, but simply due to the fact that our goal should be to be true craftsmen in everything that we do.
Coding is the closest thing we have to magic in this world, and I feel honored to be able to work with it on a daily basis along with being able to teach others how to do the same. And with that in mind it should motivate us to have a clearly defined goal of being excellent at our craft.
One of my all-time favorite baseball players was Joe DiMaggio and he had a great quote that I think is very applicable to developers. He said:
> There is always some kid who may be seeing me for the first time. I owe him my best.
I try to apply this in all of the code projects that I do, it's easy to fall into lazy habits. However, I remind myself that someone might be looking at this project and it could be the first impression they have about me as a developer, and if I took shortcuts, even if the application works, it could reflect badly on the work that I do.
This doesn't mean that you can't make mistakes, quite the opposite actually. I'm constantly striving to become a better developer and because of that, I'm always trying to work on building features and projects that I haven't created before, which naturally leads to mistakes during the learning process. However, there is a clear distinction between mistakes that get made while you're trying to build an ambitious feature compared with project bugs that pop up due to laziness and poorly written code.
# How to write better code
So, if writing bad code is immoral, what can be done to combat it? Thankfully, we have a nice set of tools and workflows that can be implemented. Here are a few of the ones that I've found to be the most effective:
* **TDD/BDD** : Regardless of your thoughts on test- or behavior-driven development, there's no denying that if they're implemented properly it can lead to a well-constructed codebase. TDD naturally leads to following best practices such as low coupling and small methods, and with its refactoring step, I'm a huge fan of using it to ensure that an application is built the right way.
* **Continuous integration** : Assuming that you have a comprehensive automated test suite, continuous integration tools such as Codeship or Travis will make sure that code will not be pushed to production until it's passed the full test suite. I've had a number of times where Codeship has blocked a bad deploy that would've taken down a site. Additionally, it gives the development team a report on what needs to be fixed if a bad deploy is attempted.
* **Pair programming** : This is one of the most powerful tools you can use as a developer. If you're not familiar with it, pair programming is the process where you and another developer both take turns working on a project at the same time, preferably in the same room and on the same computer. When one of you is coding, the other developer is watching and giving advice or warnings. Whenever I'm building a complex feature, I will always use pair programming since it's akin to working with two brains on the same feature.
* **Continuing education** : No matter how long you've been a developer, you'll never reach a stage where the learning ends. Each day I try to learn something new, whether it's from tutorials, books, or blog posts from other programmers.
# Summary
I hope that this has been a thought-provoking chapter and will help you on your journey towards becoming a code craftsman.
# Chapter 54. Inspirational Programming Advice from Howard Roark
In this chapter, I'm going to discuss one of the quotes that I've always turned to for inspiration as a developer.
It's from Ayn Rand's book, _The Fountainhead_. The main character in the book, Howard Roark, is a skilled architect and typifies the concept of being a true craftsman, in the same way that all of us, as developers, should approach our own projects.
This quote discusses how every project is special and deserves a unique implementation, and it goes as follows:
> "Rules? Here are my rules: what can be done with one substance must never be done with another. No two materials are alike. No two sites on earth are alike. No two buildings have the same purpose. The purpose, the site, the material determine the shape. Nothing can be reasonable or beautiful unless it's made by one central idea, and the idea sets every detail. A building is alive, like a man. Its integrity is to follow its own truth, its one single theme, and to serve its own single purpose."
>
> —Howard Roark, The Fountainhead
Even though this was written in the early 1900s and was for the architecture industry, Roark's approach to craftsmanship can be applied just as easily to programming. It can be easy to fall into the trap of staying in a comfort zone and simply duplicating implementation and functionality from project to project. However, that can create two problems:
1. You don't grow as a developer. The only way to get better is to step out of your comfort zone and build features that you've never done before so you can learn new concepts.
2. Projects suffer, becoming square pegs in round holes. Each application has its own set of unique requirements and therefore, should have a custom implementation.
I hope that you found Roark's wisdom inspirational and that you can apply it to your own development projects.
# Chapter 55. Best Practices Versus Creativity as a Developer
In this chapter, I'm going to discuss how you can find the balance of best practices versus creativity as a developer.
To be honest, this was initially a difficult chapter to write, mainly because I had a hard time organizing my thoughts on the topic since it's a little abstract. I had the high-level concept of the strained relationship between best practices vs creativity as a developer, in the sense that many developers, especially the ones new to coding who will fall into two camps:
* Those attempting to follow standardized conventions in every way, essentially duplicating code from tutorials and a language or framework's documentation page.
* Those ignoring all common practices and building applications in whatever way that makes the most sense to them (at the time).
There are pros and cons with both approaches, and like many other topics, a cross between the two is going to result in the best strategy. Let's look at the pros and cons of following best practices and ignoring the idea of being creative.
**Pros:**
* Application code will be easier for future developers to pick up and add features to since they'll know how the code is structured and where all of the methods and classes are located.
* Programs should be well-tested via unit and integration tests.
**Cons:**
* Code structure may have more of a cookie-cutter approach.
* Code may suffer from the square peg/round hole syndrome.
Before I go on, please don't misunderstand me by thinking that I'm saying that developers who follow best practices have these issues. I'm simply referring to developers who throw creativity out the door and simply try to build programs using standardized methods and attempt to copy code that they see from other developers.
Now let's take a look at the pros and cons of developers who only embrace creativity:
**Pros:**
* They have fun, this is very important since it keeps them motivated to build interesting projects. They're also able to express their unique perspectives on how applications should be structured.
* They're constantly trying new things and finding new ways to build features. This can result in learning quite a bit about the language or framework that they're using.
**Cons:**
* Programs that only rely on the developer's creativity can be nearly impossible to manage later. Even the developer who built the application may have a hard time understanding his own code if he has to go back and add new features.
* Even though programs built ignoring best practices and relying solely on the developer's creativity are fun to build at first, as the codebase grows the level of fun decreases exponentially. In fact, it's common for new developers to kill an entire project and have to start from scratch because the codebase became such a mess following a non-standardized approach.
So, if there are pros and cons to both approaches, which is the best way to go? I am a self-taught developer, and originally, I definitely fell into the second camp of building apps simply using creativity.
However, I ended up building some horrible applications, I did learn a lot about various languages through the process, so it was a beneficial strategy from a learning perspective. Over the years, since I matured as a developer, I realized that I had to find a balance between following standardized best practices and being able to add my own creative touches into a program.
It's simply ignorant to disregard industry-wide accepted best practices. Concepts such as properly structured, object-oriented code increase a project's maintainability and also makes it more efficient to add new features in the future. Some of the most brilliant minds in the world have spent the past century refining development procedures, and a good developer should build upon that cumulative knowledge.
With all that being said, there is still a place for being able to integrate your own creativity into a development project. The more skilled I've become as a developer, I've realized that I'm more creative with my code than I ever was before. When I originally started programming, my _creativity_ was really just an unorganized attempt to get features to work the way I thought they should be structured in my own mind. However, around a decade of experience has refined how I build programs, and that experience has allowed me to learn how to be more expressive with how I write applications.
The more confident I've become as a developer, the more I've been able to explore different ways of building projects and I'm having more fun now than ever I had before. I'll leave you with this thought from Sandi Metz, one of the software developers I personally admire the most and the author of the book _Practical Object-Oriented Design in Ruby_ when she described the balance between following cookie-cutter approaches vs implementing creativity in development, she said:
> "Design is not an assembly line where similarly trained workers construct identical widgets, it's a studio where like-minded artists sculpt custom applications. Design is thus an art, the art of arranging code."
I hope that this has been a helpful discussion and will help you find the balance of best practices versus creativity as a developer in your own projects.
# Chapter 56. A Practical Guide to Approaching Project Development
One of my favorite parts of being a teacher is interacting with students. And this chapter will focus on answering a viewer's question regarding strategies to **approaching project development**.
# Student question
The following is a letter I received from Christian, a developer from Germany:
> Starting web development a year ago, I'm currently building my first own commercial product (a classified ad site) using Rails. I really would like to hear your advice on how you approach building features on apps, as I have the following problems:
>
> I can't really plan a feature from start to end because I always think I'm missing something important
>
> Procrastination and being afraid to make errors that will be costly to correct in the future
>
> How do you approach building features when you don't know the scope exactly?
>
> What`s your process, how do you get unstuck?
>
> Best regards,
>
> Christian
# Strategies to approaching project development
In this letter, each of the questions revolve around having a strategy for approaching project development. And in this chapter, I'm going to walk through an answer for each question.
## Planning a feature from start to end
The first question from our developer Christian asks:
> "How do you approach building features when you don't know the scope exactly?"
This is a great question. As helpful as it would be to have a clearly defined scope from the beginning of the project, it rarely ever happens. And even if you were given a perfectly crafted project scope, it would most likely change during some stage of development. This would essentially render the scope pointless anyways. In our developer's (Christian) email, he said he was building out a classified ads site, so we'll use that as an example case study throughout this chapter.
## Moving from requirements to stories
One of the statements made was that you don't feel comfortable planning features from start to end because you think you're missing something. This is a common issue when you approach an application built based on a set of requirements. You probably tried putting together a list of requirements, such as:
* A user should be able to register
* Records can only be edited by the user that created the posting
If you take this approach, it's natural to feel like you're missing a feature, especially if you're new to development. Instead of using requirements, I'd recommend creating user stories. I'll give you one from a classified's app I built last year:
_A user logs in to the application. From there, she sees buttons for creating new posts or editing ones that she has created before. She only sees posts that she personally created and she can't access this page without logging in to the app. In addition to seeing her posts, she can click on a post to review all the responses from users to that post._
Notice how much more practical this is than simply listing off requirements? Your app should have dozens of user stories that contain all the initial functionality that you want to build.
## Starting with a base case
Now that your app has a nice set of user stories, how detailed should they be? Honestly, I would recommend that you keep the functionality as basic as possible. When I'm building applications, I don't even include all of the parameters that I know I'll need later.
When it comes to approaching project development I take a base case strategy. This means that I drill down a feature to its most basic component. Getting back to your classified ad app, let's say, it has a story such as:
_When a user is on the page to create a new listing, she can add a title, a description, and up to 5 pictures. From there she can click on save to create the post, which redirects her to the post show page._
If you try to build that entire feature it would be very intimidating. So, my approach would be to first simply create a form page. From there I would only add in the ability to create a post with a title. The description can be added easily in the future.
Therefore, I don't see any point in wasting time on it in the beginning. I would completely ignore the picture uploading functionality in the beginning since that will require using tools such as nested attributes. So, after a user can create a post with a title I can circle back and add each of the other components one at a time. Taking this approach makes the entire process less intimidating.
## Fear of the unknown
In the email, he mentioned being afraid of missing something important. This fear of the unknown is completely natural and let me say: you will be missing something important. Creating user stories should help to catch the critical items. However, I promise there will always be components that initially fall through the cracks. However, don't let that scare you off, you can add new features in later.
From my personal experience, I remember one time where I was building an enterprise application and completely forgot to give managers the ability to view posts they were supposed to approve! However, after I realized that I left out a key feature I could add it in and the client was happy.
So, don't let the fear of missing a key feature stop you from building. Remember the initial Facebook developer strategy.
### Moving fast and breaking things
After becoming a public company, Facebook has moved away from approaching project development in this way slightly. However, when it comes to building applications it's still a strategy that I embrace. And I've discovered that it leads to getting more done. Additionally, moving fast and breaking things will also help remedy your second problem.
## Battling procrastination
I greatly appreciate Christian's candor (which was a huge reason why I decided to give him a 1,500-word response instead of pointing him to some other resources). In the email, he admits to struggling with procrastination. And I'm glad that he did because procrastination is something that every developer I've ever known, including myself, have to fight against.
One of the top tools I use to battle procrastination is thinking small. We, as developers, naturally tend to push challenging features away. Instead, we like to focus on working with components that we're already comfortable with. That's natural for everyone, however it's not good. And I have to remind myself of this fact daily.
In his book _Deep Work_ , Cal Newport discusses how deep work (the type of work that takes someone from good to great) is not fun. In fact, when researching top performers, Newport found that 100% of the individuals did not enjoy working on challenging tasks. So, when it comes to approaching project development, please understand that the scariest features might be your greatest catalyst for improving as a programmer.
## Small, practical steps
So, my personal recommendation is that you embrace the difficult features, but take an incremental approach. For example, imagine that you're intimidated to build in the ability to let users upload multiple images per listing in your classified app. This feature can intimidate even experienced developers. You'll need to use nested attributes and incorporate a number of JavaScript elements to allow for dynamic behavior.
This is the type of feature that could lead you to procrastinate since it's a bit on the scary side. However, I'd recommend that you tackle the feature right away. You can write down a strategy for how you're going to build the component, such as:
1. Integrate nested attributes for posts.
2. Build a JavaScript script that can dynamically create new file upload elements.
3. Hard code some image URLs in a sample record in the database.
4. Finally connect the system to the storage engine.
And then guess what, you're done! Notice how much less intimidating the feature seems when you break it down into smaller steps? This is the key to fighting procrastination.
## Getting unstuck
When it comes to development there are a number of ways developers can get stuck.
### Application bugs
If you find yourself getting stuck on a bug, I highly recommend isolating the feature that isn't working. Too many times developers attempt to fix a component while still trying to keep the rest of the system functioning properly.
When it comes to debugging, I throw best practices and form out the window. Everything is on the table when it comes to fixing a bug. For example, if data isn't showing up properly on a page, put a database query in the view template. Gasp! Don't worry, after you've discovered what the bug is, you can then immediately refactor the code to conform to best practices.
Other techniques I find very helpful when it comes to getting unstuck is using the Rails console to run scripts and working with debugging tools such as Pry. These tools allow you to isolate the issue and focus on the problem instead of letting the rest of the application get in the way.
### Messages over models
Lastly, a key differentiator I've discovered between good and great developers is in the way they look at application development. A good developer can look at a program or feature and start by listing out all the models, their attributes and relationships.
However, great developers first focus on the messages that will be sent from class to class. This is a completely different way to think about development. Instead of looking at classes like static object blueprints, it forces you to think about the actual behavior of the classes.
This isn't an idea I came up with. The esteemed developer Sandi Metz described this concept the best when she said:
> "You don't send messages because you have objects, you have objects because you send messages."
# Chapter 57. How to Practice Programming Techniques and Improve as a Developer?
Whether you are new to programming or have been at it for years, practice is important. The more you practice your programming skills, the better you will be. You have various options to practice programming techniques. These options will help you brush up on your skills and continually improve as a developer.
# Engaging in pair programming
Programming doesn't have to be a solitary activity. Instead of taking in on by yourself, engage in pair programming. Since pair programming has people working together from a single computer, it is a great way to learn different strategies for tackling problems and approaching the process. This type of practice will quickly make you a better programmer, as long as you choose a good partner.
# Utilizing open source software
Open source software is a great way to practice your programming techniques. Start by reading code from various open source projects. This will help you understand how the programmers managed to create such a successful project.
Then, participate in various open source projects. As you work, people will give you immediate feedback. It might be hard to hear the criticism from time to time, but it will help you fine-tune your skills, which will make you a better programmer.
# Visiting the DailyProgrammer subreddit on Reddit
Reddit is a community full of people who share ideas and help one another. You can get in on the action with the DailyProgrammer subreddit (<https://www.reddit.com/r/dailyprogrammer/>). This subreddit posts three programming challenges each week. The first challenge is relatively easy, and then they increase in difficulty. The community reviews the solutions and provides feedback. Use this subreddit to improve your skills while having some fun.
# Taking online courses
Sometimes you don't need to go back to school to develop your programming skills. **Massive Open Online Courses** ( **MOOCs** ) are an excellent way to brush up on your skills. You can learn at your own pace and practice the techniques that you need to work on without getting rushed. Best of all, you can get feedback during the course, which will help you become a better programmer.
You can check out my _Professional Rails Course_ (<https://www.udemy.com/professional-ruby-on-rails-coding-course/>) if you want to learn how to best utilize the Rails framework.
# Code katas
The term kata was first introduced by _Dave Thomas_ in the book, _The Pragmatic Programmer_. He borrowed it from martial arts and applied it to the programming world. To code katas, you need to take a small requirement and create the code. Then do it over and over again, improving it until it is perfect. This is an easy way to practice coding while making your code better.
# Summary
Don't make the mistake of thinking that you don't need to practice programming. You should practice as often as possible. You can leverage these recommendations to practice programming techniques to improve your skills so that you can take your career or your hobby to the next level.
# Chapter 58. What Does It Take to Become a Great Developer?
Whether you've been programming for years or if you're just now learning how to code, it's natural to ask yourself: what does it take to become a great developer?
I'm going to start off by saying that there is no right or wrong answer to this question. If you ask 100 experienced software engineers this question, you'll get 100 different responses. The reason why there's no clear-cut answer is because development is truly an art. Therefore, asking this question about programming is similar to asking what makes a great artist.
# Tips for becoming a great developer
In preparation for this chapter, I asked various developers, I read blog posts, and I listened to a number of podcasts discussing the topic. As I expected, the components of becoming a great programming are extensive.
In this chapter, I want to give an overview of the processes and requirements that I've found the most effective. The following are six traits that encapsulate the key characteristics found among great developers. I've also included some practical strategies for working with each of these attributes on a regular basis.
## Working through difficult features
Starting off with one of the most challenging traits, I've found that the only way I improve as a developer is to work through challenging concepts.
I find it disturbingly easy to fall into a routine where I only perform the same tasks again and again. I've been working as a developer for a number of years and I therefore have a nice arsenal of tools and features that I'm comfortable building.
However, I've discovered that if I simply keep building features that I'm already comfortable creating, I won't grow as a developer. It's only when I bear down and dedicate myself to work through a difficult task that I've never performed before that I become better myself.
Having the requirement of working through difficult practice isn't a concept related solely to development. The book _Peak_ by _Anders Ericsson_ and _Robert Pool_ researched peak performers in music, athletics, and essentially every other skilled profession. The results of the research revealed that individuals only show improvement when working through challenging concepts. This means that if concert violists played the same music day after day and never challenged themselves, their skill would stagnate.
The same concept holds true for developers. If you want to become a great developer, you need to work through difficult topics constantly. If you don't know where to start with finding challenging features to build, visit some of your favorite websites. You could look at Twitter, Airbnb, or Pinterest. From there you can compile a list of advanced features that you've never built before. Examples would be components such as: infinite scrolling, asynchronous notifications, or multi page authentication.
## Community contribution
With the growth of the programming industry, the open source community has expanded exponentially. The most popular languages and frameworks in the world, such as Python and Ruby, were created not by corporations, but by programmers interested in the common good.
Depending on your level of experience, community contributions will vary pretty widely. If you're a senior-level engineer, you could build an open source code library or build a feature for a programming language. However, even if you barely have any experience, you can still contribute. New developers can assist other individuals who are just starting to code.
As great as it is to give back to the developer community, there are also significant benefits to contributing. If you're building a code library that other developers will see, you'll most likely be very careful to ensure that the codebase is properly tested and functions properly. This type of development will make you an even better programmer and will help you in the long run.
## Artistry
When it comes to development, it's easy to get caught up in the day-to-day minutia of a project and forget that, at its core, programming is an art. For code to be artistic, it must be elegant, and for it to be elegant, it must be simple. Some of the best projects that I've worked on ended up having the most straightforward codebases. However, writing simple code is not as easy as you may think. Sandi Metz said this about simple code:
> "Novice programmers don't yet have the skills to write simple code."
Einstein said this about simplicity:
> "If you can't explain it to a six year old, you don't understand it yourself."
This may seem like an odd concept. However, if you've ever attempted to build a complex project that maintained an easy-to-follow code design you know it to be true. The more you improve as a developer, the most straightforward your work should be.
## Craftsmanship
Craftsmanship is closely related to artistry. However, there is an important distinction. When you're a craftsman you truly take pride in your work.
Over the years I've met all kinds of developers—from programmers who simply treated each project like a widget on an assembly line, to developers who made sure that every code file they worked on looked like a piece of art.
Personally, I've found a cross between the two concepts to be the most effective. Like many other concepts, craftsman is not isolated to programming. Growing up, my Dad, who was a Major League baseball player and is now a coach, always taught me to have what he called a Spirit of Excellence. This meant that no matter what I did or what I was working on, I had to take pride in it. He would tell me that if I was taking the time to perform a task, I might as well do it properly.
While I feel that I take pride in my work, craftsmanship is one of the concepts that I struggle with the most. I find this principle challenging because it can be difficult to find the balance between well-written code and perfect code. As the saying goes, _perfection is the enemy of great_. Therefore, it's important to ensure that you work hard to properly design your codebase.
However, don't pressurize yourself to achieve perfection. It's also important to have the mindset that no project is ever truly completed. This means that if you attempt to achieve perfection you'll constantly be frustrated. Mainly, due to the fact that you will never reach a stage where your codebase will ever be considered done.
### Steve Jobs's craftsmanship
When it comes to craftsmanship, few have taken the same level of pride in their work as Steve Jobs. This is what he had to say about craftsmanship:
> "When you're a carpenter making a beautiful chest of drawers, you're not going to use a piece of plywood on the back, even though it faces the wall and nobody will ever see it. You'll know it's there, so you're going to use a beautiful piece of wood on the back. For you to sleep well at night, the aesthetic, the quality, has to be carried all the way through."
>
> —Steve Jobs
## Adapting to change
If you've worked on any real-world code projects you can attest that there is only one true constant: change. Great developers set themselves apart from novices by how they adjust to changing requirements for an application. There are two ways that new coders struggle with change:
* _No flexibility with the code design_. This means that when a new requirement is added to the project, they will need to completely reconfigure the code to allow for the additional functionality.
* _Planning for the wrong future_. A developer may have developed a mental model of what the end project will look like, however that estimation rarely matches reality. Imagine that you're building an accounting application and you think the client is going to eventually ask for the system to be completely project based. You will make design decisions based on the workflow hierarchy that you have in your mind. However, if you're wrong, you will be forced to reconfigure the entire application.
Both these pitfalls are normal to come across on a coding journey. However, a great developer finds the balance between no design and premature design.
By building well-constructed codebases, the great programmer writes modules that have flexible interfaces that can adapt to change. They also understand that project requirements change and that the code they write has low coupling.
This means that changes to one feature in the application should have little to no impact on other parts of the program. For example, back with the accounting application, if a change is required to the payroll module, it shouldn't require you to rewrite the personnel management feature.
## Tireless learning
Lastly, in answering the question: "What does it take to become a great programmer?" I'm going to discuss the importance of tireless learning.
One of the most important factors in reaching your development goals is having a thirst for knowledge. Thankfully, you have 100% control over this requirement. Regardless of how much experience you have as a programmer, you won't ever reach a stage where you should stop learning. There will always be improved processes, new frameworks, and new languages to learn.
I've asked some senior developers that I work with how they organize their learning methods. They gave the following recommendations:
* _Learn one new language or framework each year_. This should also mean that you're building a production application during that year. It's easy to follow tutorials and build hello world applications. However, when you create a real-world program you'll be forced to work through challenging constructs.
* _Read multiple books daily_. I personally have over a dozen books that I read daily related to development. In fact, many of the topics that I discuss in this book were informed directly by the things I was reading at the time.
* _Follow advanced tutorials_. Many of the developers that I work with admitted that they prefer to learn new coding techniques by reading blogs from other programmers. There are a number of guides available online that you can go through that will teach you how to build advanced features into your applications.
* _Subscribe to newsletters_. I subscribe to a number of newsletters that are sent to me each week. This includes newsletters on Ruby, Rails, and JavaScript. These types of newsletters are a great way to stay up to date with changes in a language. They curate some of the best blog posts and tutorials from around the web.
# Summary
I hope that this has been a helpful chapter and will help you answer the question, "What does it take to become a great programmer?"
# Chapter 59. How to Stay Sharp as a Developer?
When I think of the concept of **staying sharp** , images of focused skill come to my mind. From a programming perspective, a sharp developer is one who feels confident working with challenging projects and can calmly adjust to changing requirements.
I've mentioned in previous chapters that developers never remain at the same skill level. As a programmer, you're either improving or losing your expertise, there's no middle ground. So how can you stay sharp as a developer?
# Tips to stay sharp as a developer
Here are my five great tips to staying sharp as a developer. If you follow these, you'll always feel ready to focus on the next goal or challenge in your career.
## #1 – coding exercises
First on my list are coding exercises. At the end of the day, nothing is going to help you improve as a programmer as diving straight into the code.
You may think that working on work or hobby projects are enough to keep your skills sharp. However, it's been my experience that many of the projects I manage for work don't test my skill as a developer. There are exceptions of course, but much of the coding I do on work projects revolve around application configuration as opposed to algorithm design.
Growing up, I could watch the baseball players that my Dad coached. These were Major League athletes who were at the peak of their profession. I still remember how they stayed sharp as hitters. Playing in games was not how they improved, games were where they showed off their skill. Instead they became better players by going through drills and exercises that focused on improving specific components of their game.
In the same way, we, as developers, need to dedicate time on improving specific elements of our coding techniques.
This image is a GitHub gist of some coding exercises (<https://gist.github.com/jordanhudgens/8033986>) that I go through regularly. Above each code snippet is a task, and my job is to place the implementation code below the objective. These programming exercises force me to continually refine my skill and find new and better solutions to complex problems.
### Example coding exercises
Some example questions could include:
* _Remove strings from an array that start with "system."_ : This requires me to work with the array data structure, integrate a Regular Expression matcher, and know how to remove selected elements.
* _Convert an array of strings into a hash that has the string as the key and value as the string's length._ : This exercise forces me to understand how the Hash data structure functions, how to use the enumerable `Map` method, and work with blocks.
And the list goes on and on. I'm continually adding new problems to solve and I try to spend some time each day. Around 30 minutes to an hour working through these exercises. If you are looking for a great list of programming problems check out Project Euler (<https://projecteuler.net/>). There you'll find hundreds of great challenges that you can work through.
## #2 – teaching others to code
Next on the list is to teach others how to code. I was homeschooled growing up, and at around age 12 I had to start teaching myself algebra. I initially struggled with learning new concepts and I was getting frustrated. My Mom realized that when I studied by myself I had a hard time understanding what I was reading. But it was when I explained the lesson to her that my comprehension skyrocketed.
She was teaching all of my siblings, so she couldn't be by my side all day. So instead, she got one of my sisters' dolls and sat it next to me at the table. She instructed me to read the lesson plan and then explain it to the doll.
At first, I thought it was the dumbest idea I'd ever heard. Explaining algebra to a doll, aptly named _Big Dolly_ due to her size, seemed akin to a homeless person talking to himself at a bus stop. So, I did what every 12-year old would do and I ignored the advice.... Until I went through my next lesson and realized that I had no idea what I just read. So, after staring at the doll for a few awkward seconds I started to explain the concept to her. Shockingly, by walking through the lesson with the doll I started to understand it! And no one was more surprised than myself. So, _Big Dolly_ helped get me through algebra, trig, and calculus.
### How does this apply to development?
So how does my weird story about teaching a doll algebra apply to staying sharp as a developer? Well, when I started to learn programming I didn't pick it up right away. In fact, I really struggled with how to build applications. But then I remembered back to my high school math era. But instead of bringing Big Dolly back from the attic I started creating programming tutorials.
It was through teaching others that I started to understand development better than I ever have. In fact, the origins of _DevCamp_ and _CronDose_ can be traced back to my desire to improve by own skills by teaching others. So, if you're looking to learn coding or to improve as a developer, I highly recommend that you teach others and your own expertise will grow.
## #3 – reading
Next on the list of tips for staying sharp as a developer is reading. I have a membership to Safari Books Online, which is essentially a Netflix for developers. Through that membership I have access to thousands of coding books. On my daily to-do list, I have around 7-8 books that I go through.
Sometimes I read a few small sections and other times I'll read a few chapters, depending on how much time I have. Going through development books has helped me continually refine my skills as a programmer and I'm constantly on the hunt for new great books to go through.
## #4 – newsletters
Fourth on the list are development newsletters. I try to keep my time very focused. Therefore, I limit the number of newsletters that I subscribe to so that I can dedicate time to reviewing each of them when they get published.
Some of my favorites are the newsletters from: _thoughtbot_ and _Codeship_. These types of resources contain comprehensive guides that will help give you a unique perspective on development and how to implement new features.
## #5 – tutorials
Last on the list of tips for staying sharp as a developer are tutorials. Whenever I'm learning a new language or framework tutorials are one of my favorite resources for studying.
Screencasts are the closest you can get to having an instructor in the room with you. Providing you with step by step guides that you can follow to build applications from scratch. Thankfully, with the growth of the online educational industry there are tutorials for essentially everything that you want to learn.
# Summary
I hope that this has been a helpful set of tips that will help you stay sharp as a developer, and good luck with the coding!
# Chapter 60. Developer Resume Tips – How to Create an Effective Resume?
Over the years, I've heard programmers say that they only need a resume if they want a normal job. However, it's been my experience that a resume is required for traditional job interviews, freelance clients, and even raising money for a startup. With that in mind I've put together this collection of **developer resume tips**.
# Developer resume tips
I've been on both sides of the hiring process. I have been in the place of sending out resumes to hundreds of companies, hoping for a response. And I've also been on the receiving end where I was sent countless resumes from applicants.
The list I've developed has been fine-tuned throughout the years. With the strategies coming from the resumes that resonated the most with me along with the elements that worked best when I sent them to potential employers. In this chapter, I've compiled three straightforward developer resume tips that will optimally position you with hiring managers.
## Keep it simple
Starting off the list is **keep it simple**. I have passed over many resumes that included pages of extensive descriptions and explanations of a developer's experience. If a developer sends me a resume that's over 1 page it's rare that I will take the time to go through the information.
The resume firm, Novorésumé, created a resume for Elon Musk where they condensed Musk's career down to a single page. And if Musk can have a single page resume, so can you. So, with this in mind, what pertinent should you include in your resume? A well-crafted resume will typically include information such as:
* _Your name and contact information_ : You'd be shocked how many individuals will write multiple pages listing their accomplishments but forget to leave their email, phone, and social media links.
* _Your education_ : And when I say education, I don't only mean traditional education sources such as your high school and college. The education portion of a resume should include any bootcamps or online educational institutions that you completed.
* _Your work experience_ : In your work experience section, brevity is a virtue. Hiring managers don't want to read through every little detail of every project you've been involved in. They simply want to ensure that you will be a fit for the position that they're looking to fill. That's it. It's not that complicated.
* _Your skills_ : Another common mistake I see from applications is forgetting to list out their full set of skills. As with all the other resume elements, keep this list simple as well. For example, I summarize my list of skills down to a few lines that discuss the programming languages and frameworks that I work with.
* _Your achievements_ : Lastly, you should list out any achievements or certifications that you've earned in your career.
## Keep it relatable
Next on the list is to **keep it relatable**. If you are applying for a frontend developer position, customize your resume to revolve around your frontend skills and experience. When I was sending out my resume to companies, I customized the content of the resume for each organization. If a company said that it was looking for a server-side specialist, I created a resume that highlighted my server-side expertise.
If you think that creating custom resumes is sneaky you'd be wrong. Hiring mangers typically decide on whether or not to contact you within a few seconds of glancing at your resume. By creating a custom resume that outlines a set of skills and experiences that fit the company's needs you are helping to make the hiring manger's job easier. So, it's a win-win scenario.
## Keep it professional
Third on the list of developer resume tips is to **keep it professional**. Companies hiring managers really could care less that you like animals or that you enjoy running marathons. You can discuss your hobbies during the interview process. However, on a resume, hobbies take up precious page real estate that can be utilized by listing out additional skills or experiences.
# Summary
In summary, writing a resume is truly an art. Resumes should be succinct, customized to the job position, and clearly describe why you are the best person to fill a position. If you have never written a resume before, I highly recommend that you work with a service or individual that can help you with the process. Services such as Novorésumé are great for this type of work and you can also hire an experienced resume writer from sites such as Upwork for under $100. Considering that a well written resume can make the difference between getting a job or not, I think it's a wise investment.
# Chapter 61. Developer Salary Negotiation Strategies
Talking about money is a sensitive subject for many individuals, and when it comes to negotiating how much you'll be paid it also comes with the added stress of knowing that if you ask for too much you may not get the job and if you ask for too little it could negatively affect your lifestyle. With that in mind I've put together a list of developer salary negotiation tips.
# Knowing your skill set
This may seem like common sense, however it's vital that you know and can articulate your full set of skills since this is going to be one of the main factors that dictate your salary.
You can start with listing out the programming languages you know, the frameworks that you've used, and put together a portfolio that showcases your expertise. As an example, if you're a full stack developer who also has experience with data science, you will be more a more valuable asset to companies that require a unique skill set like yours.
# Knowing the industry
Over the years I've been fortunate to work as a VP of Engineer and a software Director in the oil and gas industry. The oil and gas industry has historically had a difficult time attracting software developers compared with other sectors, and because of that developers are able to command a premium salary. I would make a much different salary if I applied to work for the automobile manufacturing industry compared with the oil and gas space.
Therefore it's important to understand what industries pay for developers and not to simply assume that the same skill set is paid the same amount across all sectors.
# Knowing the organization
No matter what your skill set is and what industry you're working in, no factor will determine your salary as much as the organization itself. If you're applying to work for a bootstrapped startup you'll be paid significantly less compared with a startup that just finished raising $20 million of venture capital.
It's been my experience that small to medium sized companies, with around 500-1000 employees pay the most. If you are applying to work with a bootstrapped startup, they may be willing to negotiate with stock options which could eventually lead to a much larger payday than any salary would, so that is also important to keep in mind.
# Researching salary rates
With all of that in mind, how can you research salary rates? You can always simply Google _software developer salary_ and then name the industry that you are interested in. Typically, career sites with job adverts will give you a huge indicator. However, I typically like to use _Glassdoor_ , which I've found to have the most accurate salary rates.
You can also have Glassdoor filter by the location that you want to work in. One key item to keep in mind is to test out multiple job types. For example, when I search for _web developer_ jobs in Scottsdale, AZ, it showed an average salary of $65,000, however when I looked up 'software developer' jobs in Scottsdale it returned an average salary of $78,000. That's a pretty big salary bump for a single word difference, so make sure that you check out all of the potential job types that you're interested in.
Another interesting option is to watch some of the yearly developer surveys that some large sites conduct, such as the annual developer survey at Stack Overflow. These kinds of surveys will break the industry down into various sectors and request information about typical salary rates. While you can't take surveys like this as facts, they are an excellent indicator.
I hope that this guide has given you a system for negotiating your next salary.
# Chapter 62. Best Questions to Ask During a Job Interview
In this chapter, I'm going to discuss the **best questions to ask during a job interview** , and I will also discuss the other side of the spectrum and list out some key interview questions to avoid at all costs.
If you haven't been through many job interviews for a while, or haven't really ever prepared for one before, you may have the thought:
_"I thought the interview was about them asking me questions."_
However, one very important, and many times overlooked facet of a good interview is asking strategic questions of the interviewer. In fact, Forbes researchers have outlined three goals that your questions should achieve:
* Make sure the interviewer has no reservations about you
* Demonstrate your interest in the employer
* Find out if you feel the employer is the right fit for you
Interviewers like being asked questions. The questions you ask can reveal quite a bit about yourself, good and bad, therefore it's critical to ask the right questions. If you don't ask an interviewer questions, he or she may assume that you don't really care about the job itself and you're simply looking to make enough money to pay your rent.
Interviewers want to find candidates excited about working with their company. Remember that if you're hired, you are going to be a reflection of the interviewer and will help or hurt their reputation. Anytime that I've hired an employee that turned out to be bad for an organization, the management has approached me and asked why I hired them and how did I miss their shortcomings.
I kid you not, I hired an individual for a job around five years ago that turned out to be a horrible employee and our CEO still gives me a hard time about the hire half a decade later! So, make sure to take all of that into account when you're interviewing for a position.
The following is a list of the best questions to ask during a job interview along with rationales on why they're good questions to ask. I've picked these up through my years as a manager for several software companies and through researching the topic.
# Best questions to ask during a job interview
_"How is performance measured for this position?"_
This is probably my favorite question to be asked. It shows that the individual is not only interested in the job, but also wants to have an understanding of what it takes to be successful. A key component of this question is also that the question focuses on how the company measures performance. This will give you a good idea of how data-driven the organization is and focuses on the key metrics that are important to the company.
_"What are some specific challenges that I will be tasked with?"_
Asking this question will show that you are not naive and that you understand that the job will have challenges and that you want to prepare for them in advance.
_"Are there any responsibilities with the position that were not mentioned in the job posting?"_
This type of question will tell the interviewer that you're savvy and experienced enough to know that 100% of the requirements don't always make it to the job website. For example, the job may be a software development job, however they may also want you to perform search engine optimization on the web application. This question not only positions you well with the interviewer but also will help you understand the full set of roles and responsibilities that the position will entail.
_"What is the corporate culture like?"_
I like this question because it will give you a feel for how employees interact with each other and management, it will also let the interviewer know that you aren't purely looking for a 9-5 job, you are interested in working with the team and fitting in.
_"On average, how long do employees stay with the organization?"_
Similar to the corporate culture question, this will subtly let the interviewer know that you are not looking for a short stint at the organization, but that you are looking for a long-term relationship with the company.
The answer to this question will also help provide you with the understanding of how tolerant the company is with regard to keeping employees. If the average employee has been with the company for only a few years there may be some issues causing the rapid turnover. Whereas, if employees stay on for over a decade, it's a good sign that the organization is a great place to work.
_"Do you have any hesitations about my qualifications or experience?"_
This is a bold question and you may or may not want to ask it depending on your confidence level. However, it will let the interviewer know that you're not afraid to ask tough questions and that you're willing to hear constructive criticism. I've personally never been asked this question by an applicant but I would admire anyone that would be willing to ask it.
# Poor questions to ask during a job interview
You will want to stay away from questions that appear that you want to get out of work such as asking about the amount of paid time off, tiers of vacation days, and so on. It's important to know these parts of the job, however it will reflect better on you if you instead ask questions such as:
_"What types of benefits are associated with the position?"_
This question will get the interviewer to give answers to the PTO and vacation time without you coming across like someone who's already trying to see when they won't have to work, but still get paid.
Typically, you will also want to not ask questions that start with _why_ , because _why_ questions will immediately make the interviewer defensive and will give the conversation a feeling of being confrontational. These types of questions can usually be changed to start with _how_ , which will tell the interviewer that you are wanting to know more about the company instead of accusing the company of doing something wrong.
An example of this would be _"Why does the company pay by check instead of direct deposit?"_ This makes it seem like you're saying that the organization isn't staying up with modern payroll procedures (which may or may not be the case), however nothing good can come from this type of question.
Hopefully this gives you an idea of what questions not to ask, but just in case a few more case studies would help, here are some more questions that you should avoid:
_"Is telecommuting a possibility for the position?"_
If it was they would've already mentioned it, if telecommuting is your top goal you should consider freelancing.
_"How much does the position pay?"_
You should already have a decent idea of how much the job pays if you're interviewing for it. You can worry about the wage after it's been offered to you, it's never a good thing when the interviewer thinks that you're top priority is how much money you're going to make because they'll assume that you will leave the company if another organization offers you a modest raise.
_"What type of hours would I be expected to work?"_
Interviewers and managers hate this question! Based on the job you should already know this. However, you can reframe this question by asking something like _"What does a typical day for this position look like?"_.
Lastly, never ask _"How did I do?"_ or _"Do I have the job?"_. Asking how you did sounds like you just finished a spelling test in 3rd grade and doesn't reflect much confidence. And if you had the job they would have told you. Wait and follow up with the interviewer in a week and you'll be considered a mature and experienced candidate.
# Summary
I hope that this has been a helpful guide and will help you put together a strategy of the best questions to ask during a job interview. Many aspects of an interview aren't known going into it, however you will always have the ability to control the questions you ask, so it is good to put together a list of questions to ask so you can maintain some level of control over the interview process.
# Chapter 63. Answering in an Impossible Interview
# Questions
How would you like to be asked this question in an interview:
_"How long would it take to sort 1 trillion numbers?"_
Or
_"How many planes are there in the sky at a given moment?"_
If that doesn't sound like your idea of fun you're not alone and many organizations have stopped asking impossible or seemingly impossible questions; however, there are still interviewers who like to ask them, so I thought it would be helpful to discuss.
First and foremost, if you get asked one of these questions, the interviewer's top goal is not seeing if you know the answer or not, instead they are looking to observe your problem-solving skills. Typically, the best approach is to take a systematic strategy so you can show that you have an organized thought process.
# Answering impossible interview questions – case studies
Let's take the first question as an example: _"How long would it take to sort 1 trillion numbers?"_ That seems like a challenging task until you establish a base case, such as: _"How long would it take to sort 10 numbers?"_
If you know your algorithms, you would most likely choose to use a sorting algorithm, such as _Quicksort_ , since it has an average sorting time of _O(n lg n)_ , which would be _O(100 lg 100)_ after swapping out _n_ for the total number of integers. So, it's actually trivial to sort 1 trillion integers, since it would be _O(1,000,000,000,000 lg 1,000,000,000,000)_.
In this question, the interviewer is first wanting to ensure that you know popular algorithms and that you have a clear understanding on which algorithm would be the right fit for the task.
For _"How many planes are there in the sky at a given moment?"_ this is a guesstimate question and the actual answer doesn't matter at all, instead the interviewer wants to see how you walk-through an analytical problem and also how detailed oriented you are.
For this I would start by asking the interviewer filtering questions, such as: all the planes in the world or only in the US? Does this only include commercial planes or private jets and military aircraft? And questions that show that you know how to properly think through each of the parameters you would need to know to properly answer the question.
From there break down the problem into subproblems; for instance, if the interviewer says that they only want the list of commercial planes in the air in the US, then you can start to create an estimate on how many flights leave an airport each hour, set an average duration of the flights, and so on. Then you can multiply that base case by the estimated number of airports in the US and you will have your guesstimate, and more importantly you will show the interviewer that you have thought through the problem.
I hope that this guide has helped give you a strategy on **answering impossible interview questions** and good luck with the job hunt!
# Chapter 64. Greatest Weakness Answers for Coding Interviews
What's your greatest weakness? To be 100% honest, I'm not a huge fan of this question since it's a lose-lose question: if the individual being interviewed is completely honest, they probably won't get the job, and if they give a flat out lie it will be evident very quickly.
However, throughout the years I have had several managers ask me this question, so it's important to have a well thought out answer ready. Since I've been on both sides of the interview chair, I have put together a list of the worst ways to answer this question and then some of the best answers.
# Bad answers to your greatest weakness
So first of all, these are the kinds of answers you should avoid to the infamous question about what is your great weakness when you're in that coding interview:
* _Hard time saying no_ : This will say that you are weak and will take on too many projects, typically resulting in poor performance.
* _Can have a hard time staying on a single task_ : This means you should probably either learn how to focus or that you are supposed to be an entrepreneur, working for a company usually means long hours on long, repetitive tasks, and if someone is bouncing from project to project they're not going to be a good asset to the team.
* _Can be arrogant_ : I had a professor in grad school, Dr. Richard Gelfond, who is one of the most brilliant computer scientists I've ever known, who said this about arrogance: _"Being arrogant occasionally doesn't matter, but it's never good"_. An arrogant employee won't be able to work well with others and will have a hard time taking correction or learning anything new... because they already know it all.
# Good answers to your greatest weakness
Now let's see some great answers to the question about what is your great weakness when you're in that interview:
* _Can be slow to take action_ : This is an interesting answer because it could be a bad answer if you leave it at this since it might tell the interviewer that you are timid, which would be a poor character trait. However, if you follow it up by saying that you are very detailed and that before you start a task you want to have a clear strategy for how to move forward, that is an employee I want to have on my team!
* _Can be overly critical on myself_ : This will tell the interviewer that you care about your work and that doing it well is very important to you. Make sure to follow up with how you're working on yourself so that you still are working on being a craftsman at your work but to not get on yourself for irrelevant reasons.
* _Can be quiet_ : This is one of my favorite things to hear, assuming that the individual follows it up by saying that they can be quiet because he or she thinks that it's important to listen to all of the details of a project or challenge before giving an answer. This tells me that the individual is going to be detail oriented and won't miss key items because they were too busy thinking what they were going to say next.
I hope that these strategies help you answer the question of: What's your greatest weakness? And good luck with the interview!
# Chapter 65. Enterprise Software Job Strategy and Guide
If you're a developer and have worked with startups or freelance clients, you'll find that the world of enterprise software jobs is quite different, in both good and challenging ways.
Some of the pros to working in the **enterprise software industry** is that large organizations typically are stable, have systems in place for development, and allow you to specialize on a specific piece of functionality instead of having to cover the full range of software features.
Some of the challenges that are unique to enterprise development are that there is typically quite a bit of red tape for developers. If you're used to being able to grab any code library and stick it in your application you'll find that enterprises are pretty picky about what outside libraries you bring in. For example, I was just talking with a enterprise software developer a few days ago who expressed how frustrating it was that it took two months for his company to give him permission to use the jQuery library, which is one of the most commonly used user interface libraries in the application space.
So what types of questions should be prepared for when applying for an enterprise software job?
* Questions about how well you work with **formal processes**. Depending on the company's level of formality you'll most likely need to explain how you have worked with different project management processes such as Scrum, Extreme Programming, and processes such as that.
* At large enterprises software bugs can cause millions of dollars in damages, so the **testing systems** are typically quite thorough. With that in mind, make sure that you are prepared to answer questions related to unit testing and behavior-driven development.
* Considering that we're living in a **Microsoft-based enterprise world** , having a solid knowledge of how well you understand ways to integrate with Windows servers, Active Directory, and tools such as SharePoint will be very important. I've personally been asked how I would integrate Active Directory **single sign on** ( **SSO** ) into a Ruby on Rails application, along with how to run Rails on a Windows server during various interviews. And if you know the Rails development system you'll know that these are not standard requirements at all.
* As with most development positions, you'll also need to have a solid understanding of the **technical skills** of the position. This usually includes being able to give accurate answers to questions related to object-oriented principles and a walkthrough of popular algorithms.
The technical portion of interviews will most likely be specific to the job. For example, if you're applying for a frontend enterprise software job they probably won't ask you about Quicksort, but they will ask you about how to properly manage JavaScript callbacks.
# Summary
Hopefully, these four areas of questions will help you prepare and feel confident about applying for an enterprise software job, and good luck with the interview!
# Index
## A
* advanced features
* managing / Managing advanced features
* application development process / The process
* approaching project development
* about / Student question
* strategies / Strategies to approaching project development
* feature, planning / Planning a feature from start to end
* requisites / Moving from requirements to stories
* base case / Starting with a base case
* fear, of missing key feature / Fear of the unknown
* key feature, strategies / Moving fast and breaking things
* procrastination, battling / Battling procrastination
* practical steps / Small, practical steps
* unstuck, obtaining / Getting unstuck
* application bugs / Application bugs
* messages, over models / Messages over models
## B
* Basecamp
* about / Basecamp
* BDD / How to write better code
* blogging / Blogging
* bookkeeping options
* about / Freelance bookkeeping options
* FreshBooks / FreshBooks
* QuickBooks / QuickBooks
* NetSuite / NetSuite
* Build reporting engine / Hacking the plan
## C
* client communication
* maintaining, with system / A system to maintain proper client communication
* client conflicts
* managing, with strategies / Strategies for managing client conflicts
* clients
* obtaining / Where to find new clients
* friends and family / Where to find new clients
* network events / Where to find new clients
* outsourcing services / Where to find new clients
* obtaining, outsourcing services / The challenge in getting new clients with outsourcing services
* obtaining, as freelancer / Getting new clients as a freelancer
* proposal material / Proposal material
* constant proposals, sending / Sending out constant proposals
* result / The result
* code
* better code, writing / How to write better code
* code faster
* about / Learn how to code faster, From classical music to coding
* default mind / Our default mind
* hacking / Hacking the mind
* slowing / Slowing it down
* Beethoven / Bend it like Beethoven
* code learning
* deciding on / Should I learn to code? – a balanced look at both sides
* reference / Should I learn to code? – a balanced look at both sides
* code libraries
* Pundit / Becoming a sharp shooter with code libraries
* Devise / Becoming a sharp shooter with code libraries
* coding / Coding is hard
* coding exercises
* reference link / #1 – coding exercises
* coding interviews
* weakness / Bad answers to your greatest weakness, Good answers to your greatest weakness
* compounded learning
* about / Compounded learning
* case study / A compounded learning case study
* comprehensive study system / The comprehensive study system
* continuous integration / How to write better code
* conversational skills
* about / Conversation
* tips / Conversation tips
## D
* DailyProgrammer subreddit
* URL / Visiting the DailyProgrammer subreddit on Reddit
* visiting / Visiting the DailyProgrammer subreddit on Reddit
* deep work
* definition / Definition of deep work
* strategy, for developers / The deep work strategy for developers
* action, taking / Taking action
* distractions, removing / Removing distractions
* studying / Study hard and smart
* demonstrations
* importance / Why proper demonstrations are important
* developer
* about / Do you have to be a genius to be a developer?
* developer resume, tips
* about / Developer resume tips
* simple / Keep it simple
* relatable / Keep it relatable
* professional / Keep it professional
* developers
* significance / Following your passion – a case study
* developers, characteristics
* challenging traits, handling / Working through difficult features
* community contribution / Community contribution
* artistry / Artistry
* craftsmanship / Craftsmanship
* adapting, to change / Adapting to change
* tireless learning / Tireless learning
* developer soft skills
* about / Developer soft skills
* writing skill / Writing
* conversational skills / Conversation
* conversational skills, tips / Conversation tips
* management skill / Management
* design skill / Design
* public speaking skill / Public speaking
* public speaking skill, tips / Becoming a better public speaker
* importance / The importance of soft skills
* developer specialty
* selecting / How to choose a developer specialty?
* full stack developer / #1 – the full stack developer
* server-side developer / #2 – the server-side developer
* frontend developer / #3 – the frontend developer
* mobile developer / #4 – the mobile developer
* data scientist / #5 – the data scientist
* decision making / Making the decision
## E
* effective study practices
* case study / An effective study practices case study
* reification example / The reification example
* expert positioning / Expert positioning
## F
* fear of success
* hacking / Hacking the fear of success
* forced repetition / Forced repetition
* freelance business
* organically growing / Organically growing a freelance business
* referral requests / Referral requests
* blogging / Blogging
* expert positioning / Expert positioning
* open source contribution / Open source contribution
* social media marketing / Social media marketing
* freelance developer
* about / A clear end
* scope creep / What is scope creep?, When scope creep isn't scope creep, When scope creep goes badly
* scope creep, requisites / Based on requirements
* scope creep, approach / Based on a story
* scope creep, sign off / The sign off
* freelance portfolios, examples
* about / Examples of freelance portfolios
* social network utility, building / Social network utility
* API tool, developing / An API tool
* accounting application / An accounting application
* scheduling application, creating / A scheduling application
* frontend application, creating / A frontend application
* freelance requirement elicitation
* about / Freelance requirement elicitation
* feature, adding / How it started
* feature, building / The build
* feature, issue / The problem, Who was at fault?
* solution / A better way, Step 1, Step 2
* conclusion / A better ending
* freelancing services
* about / Freelancing services
* FreshBooks
* about / FreshBooks
* working / How it works
* features / FreshBooks additional features
* drawbacks / Weaknesses
* Frustration Zone / Frustration = skill
## G
* GitHub
* about / GitHub
## H
* hacking / Hacking procrastination
* hard way
* about / The hard way
## I
* impossible interview
* questions / Questions
* answering / Answering impossible interview questions – case studies
* case studies / Answering impossible interview questions – case studies
## J
* job interview
* best questions / Best questions to ask during a job interview
* poor questions / Poor questions to ask during a job interview
## K
* Kanban boards
* reviewing / Kanban
* katas
* coding / Code katas
* Kouros
* mental models / Mental models for the Kouros
## L
* LeanKit
* about / LeanKit
* learning curve
* about / What is the learning curve?, The developer learning curve, A unique journey
* liftoff / Liftoff
* twilight zone / The twilight zone
* zone / The zone
* learn programming / But you can learn programming
* legacy application, tips
* about / Tips for taking over a legacy application
* test suite, creating / Creating a test suite
* new features, adding via TDD / Adding new features via TDD
* specific features, breaking out into microservices / Breaking out specific features into microservices
* codebase, drying up / DRY up the codebase
* legacy scenario / The legacy scenario
* LinkedIn
* about / LinkedIn
## M
* Massive Open Online Courses (MOOCs) / Taking online courses
* memorization
* guidance / The guide to memorization
* repetition / Repetition, Smarter, not harder
* visual mental mapping / Visual mental mapping
* short-term memory, versus long-term memory / Short-term versus long-term memory
* visual mental mapping, implementing / Implementing visual mental mapping
* real-world example, taking / Taking a real-world example
* patterns, finding / Finding patterns
* copy/paste / Copy and paste is the enemy
* avoiding / Not everything has to be memorized
* mental models
* for Kouros / Mental models for the Kouros
* about / What are mental models?
* for developers / Mental models for developers
* mind works / The way the mind works
* reasoning / The reason
* smarter approach / A smarter approach
* mistakes
* creating / The secret weapon to mastery – making mistakes, Making mistakes – memory steroids
* force learning / Mistakes force learning
* kill pride / Mistakes kill pride
* multiple sessions / Multiple sessions
## N
* narrowed focus / Narrowed focus
* negative effects
* additional / Additional negative effects
* NetSuite
* about / NetSuite
## O
* online courses
* utilizing / Taking online courses
* open source contribution
* about / Open source contribution
* direct code contribution / Open source contribution
* tutorials / Open source contribution
* pre-existing libraries, contributing / Open source contribution
* open source software
* utilizing / Utilizing open source software
* outsourced web developers
* managing / A system to manage outsourced web developers
* automated testing / A system to manage outsourced web developers
* daily reports / A system to manage outsourced web developers
* applications, accessing / A system to manage outsourced web developers
## P
* pair programming / How to write better code
* engaging in / Engaging in pair programming
* perfectionism
* hacking / Hacking perfectionism
* plan
* hacking / Hacking the plan
* plateau
* learning / What is a learning plateau?
* false ceiling / False ceiling
* pomodoro technique
* using / Practical ways to use the Pomodoro Technique, Taking a break
* lifestyle, versus fads / Lifestyle versus fads
* lifestyle of productivity / A lifestyle of productivity
* implementation / Practical implementation
* PowerPoint / PowerPoint
* practical system / A practical system
* procrastination
* root causes / Root causes of procrastination
* hacking / Hacking procrastination
* issues / The problem with procrastination
* instant gratification / Instant gratification
* coding steps / Baby steps to knock out procrastination, Baby coding steps
* prodigies
* about / Are prodigies real?
* Mozart case study / The Mozart case study
* developers / Are developers born or made?
* tipping point / The tipping point
* prodigy myth / Why we love the prodigy myth
* profession
* skill, improving / Improving your skill in your current profession
* practical / Is this practical?
* professional developer / Becoming a professional developer
* developer bootcamps / Developer bootcamps
* practical / Is this practical?
* programming expertise
* degrees of / Degrees of programming expertise
* learning / Learning for fun or as a hobby
* programming language
* selecting / How to pick a programming language?
* selecting, based on job / The next job you want
* selecting, based on development specialty / Your specialty
* Project Euler
* URL / Example coding exercises
* ProWorkflow
* about / ProWorkflow
* public speaking
* about / Public speaking
* tips / Becoming a better public speaker
## Q
* quality
* versus speed / Quality versus Speed
* QuickBooks
* about / QuickBooks
## R
* Rails framework
* online reference / Taking online courses
* reading
* significance, for developers / Why is reading important for developers?, Compounded learning, Audio books are books too!
/ Reading
* reading schedule / The reading schedule
* reading system / My reading system
* real-world projects / Real-world projects
* Reddit
* DailyProgrammer subreddit, visiting / Visiting the DailyProgrammer subreddit on Reddit
* about / Visiting the DailyProgrammer subreddit on Reddit
* refactoring
* versus starting over / Starting over versus refactoring, When should you start over?
* fear factor, removing / #1 – removing the fear factor
* 80/20 principle, analyzing / #2 – analyzing the 80/20 principle, #3 – building an automated bug list
* automated bug list, building / #3 – building an automated bug list
* client, becoming / #4 – becoming the client
* architecture, changing / When should you start over?
* language/framework, moving / When should you start over?
* referral requests / Referral requests
* referrals
* about / Referrals
* reification example
* model / The reification example
* controller / The reification example
* view / The reification example
* remote desktop
* about / A remote desktop
* GoToMeeting / A remote desktop
* screen sharing / A remote desktop
* free options / A remote desktop
* reverse note taking
* about / Reverse note-taking
* benefits / Benefits of reverse note-taking
* narrowed focus / Narrowed focus
* story-based mindset / Story-based mindset
* forced repetition / Forced repetition
* running man / The running man
## S
* salary negotiation, tips
* skill set, knowing / Knowing your skill set
* industry, knowing / Knowing the industry
* organization, knowing / Knowing the organization
* salary rates, researching / Researching salary rates
* scope creep / What is scope creep?
* screencast / Screencast
* SEO, best practices
* about / SEO best practices tutorial
* relevant content / Content is king
* XML sitemap, creating / Creating an XML sitemap
* text, images, and videos, mixing / Mixing text, images, and videos
* site speed, managing / Managing your site speed
* site responsiveness, managing / Site responsiveness
* backlinks / Backlinks
* focused content / Focused content
* services
* review, to remotely demo work / Review of services to remotely demo work
* silver bullets
* issues / The problem with silver bullets
* customization / Silver bullet customization
* code libraries / Becoming a sharp shooter with code libraries
* skill plateaus
* obtaining / Getting past skill plateaus
* proper information/resources / Proper information/resources
* best practices / Best practices
* challenging / Challenging/new tasks
* about / Frustration = skill
* small bites / Small bites
* social media marketing / Social media marketing
* specialty-based mapping
* about / Specialty-based mapping
* full stack development / Specialty-based mapping
* frontend development / Specialty-based mapping
* server-side development / Specialty-based mapping
* data scientist / Specialty-based mapping
* speed
* versus quality / Quality versus Speed
* starting over
* versus refactoring / Starting over versus refactoring
* staying sharp
* tips / Tips to stay sharp as a developer
* coding exercises / #1 – coding exercises
* coding exercises, example / Example coding exercises
* code, teaching / #2 – teaching others to code, How does this apply to development?
* reading / #3 – reading
* development newsletters / #4 – newsletters
* tutorials / #5 – tutorials
* story-based mindset / Story-based mindset
* system
* studying / A system for how to study
* for decreasing task switching costs / A system for decreasing task switching costs
## T
* TDD / How to write better code
* test-driven development
* importance / Importance of test-driven development
* regression / Importance of test-driven development
* team management / Importance of test-driven development
* documentation / Importance of test-driven development
* development process, leading / Importance of test-driven development
* test-driven development (TDD) / TDD on freelance projects
* on freelance projects / TDD on freelance projects
* decision, creating / Making the decision
* choice, consideration / Giving no choice
* client, decision / Letting the client decide
* common sense, using / Using common sense
* tipping point
* for developers / Tipping point for developers
* own experience / My own experience
* doubt machine / The doubt machine
* painful process / The painful process
* (s) / The tipping point(s)"), The first tipping point, The second tipping point, The secret, The book
* top project management tools
* about / Top project management tools
* Basecamp / Basecamp
* Trello / Trello
* LeanKit / LeanKit
* ProWorkflow / ProWorkflow
* Wrike / Wrike
* GitHub / GitHub
* traditional note taking
* issues / The problem with traditional note-taking
* traditional study habits
* limitations / Why traditional study habits don't work
* Trello
* about / Trello
* tutorials / Tutorials
## U
* updates
* importance / Importance of daily updates
* client update, example / An example of client update
* client update, automating / Automating client updates
* version control / Version control to the rescue
* Upwork / The talent pool
* about / Freelancing services
* urgent client
* about / My urgent client
* firing / When to fire a client, The joy of firing a client
* employee, treating / #1 – being treated like an employee
* tyranny / #2 – tyranny of urgent
* toxic environment / #3 – toxic environment
## W
* willpower limits
* about / What are willpower limits?, Are willpower limits real?
* decision, making / How many decisions do you make each day?
* importance / Why is willpower important?
* executing / When the willpower well runs dry
* saving up / Saving up willpower
* outfit / One outfit to rule them all
* copycat / Being a copycat
* focusing / Focusing willpower
* Wrike
* about / Wrike
|
{
"redpajama_set_name": "RedPajamaBook"
}
| 5,466
|
package com.easyjf.web.interceptor;
/**
* 拦截器异常
*
* @author 大峡
*
*/
public class InterceptorException extends Exception {
private static final long serialVersionUID = -6542085034837683608L;
public InterceptorException(String info) {
super("Interceptor exception info :" + info);
}
}
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 7,912
|
I am a class 12 student(Humanities) in K V Pattom(shift 1).
That news of your cowardice on the battlefield.
I love my God.I thank my Allah.I talk to him all day. He is the greatest. By giving me this height to reach people, he has also given me great responsibilities. Peace in every home, every street, every village, every country-this is my dream. Education for every boy and every girl in the world.
To sit down on a chair and read all my books with all my friends at school is my right. To see each and every human being with a smile of happiness is my wish.
These words helped me the most to see through this brave girl. This helped me to believe in the power of one person's voice to inspire the change in the world. A life void of taking risks is not a life worth living.Fear will drive you into fear and fear and fear… If a 14 yr old girl can withstand the threat of Taliban,then youth can miraculously withstand the threat of illiteracy, poverty and deprivation.
I come from a country which was created at midnight. When I almost died, it was just after midday.One year ago, I left my home for school and never returned".
The book sets off with the prologue titled 'The day my world changed' of which the first sentence is this .
It is a unique autobiography of a 14 year old Pakistani girl, Malala Yousufzai who scared death to death while fighting for her right to education that was denied to hundreds of girls like her in Swat Valley by Taliban. She describes the story of her family since her birth.Any reader is going to be fascinated by the wisdom and realisation she holds at this age.
O Malalai of Maiwand, Rise once more to make pashtuns understand the song of honour, Your poetic words turn worlds around, I beg you, rise again.
Further,she explains Pakistan in its real sense.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 2,954
|
Depeche Mode 12" Singles Collection vinyl
by obs » Thu Aug 23, 2018 2:12 am
Has this been mentioned here?
I just saw these listed at Amazon. Seems they're a series of vinyl boxsets compiling the 12" singles for each album. First two are coming out August 31. $60+ CAD or $54 USD, even more
- includes a facsimile reproduction of the rare Flexi Disc "Sometimes I Wish I Was Dead" b/w "King of the Flies" (the Fad Gadget track as on the original release)
https://www.amazon.ca/dp/B07DQ963QD/
A Broken Frame
- original single poster reproduction (never heard/seen this before)
https://www.amazon.ca/dp/B07DQ31ZY6/
I did post in these forums years ago that they should reissue the 12"es instead of reissuing the albums again and again and again .
It would be cool if the outer jacket was the 12" artwork, and the inner was the 7" artwork. It's a shame they didn't include the tracks from the 7"es on these. I guess it would be too much, depending on how they plan on doing future boxsets. (eg. Will they combine the 12" + L12" on a single vinyl?)
Re: Depeche Mode 12" Singles Collection vinyl
by negative1 » Thu Aug 23, 2018 12:22 pm
obs wrote: Has this been mentioned here?
the problem with this is 'dreaming of me', was never a 12 inch (even though there's an unreleased extended mix).
so you're not really getting the real 12 inches, also some had alternate artwork for different regions, etc.
they're pricey, so i'll just wait until they go on sale someday.
negative1
by Jim2 » Thu Aug 30, 2018 11:39 am
I just opened my box sets.
The Sometimes I Wish I Was Dead fexi disc is thicker than the original version.
The posters, although folded in half, cry out to be framed. I will look into doing this.
The labels are reproduced like the originals except on the Leave In Silence 12", where the red ink for the titles on the labels is glaringly too bright to read.
I cannot take them home to play them until tomorrow (street date), but I'm sure they will sound wonderful.
I will purchase every one of these box sets, even though I own every original Depeche Mode 12 inch, but will stop listening to them with the Exciter release.
obs, if these sell enough copies, I'm sure there will be a 7" version forthcoming.
This is my crack.
"Things went fine, up until the start of school, the world was still okay." -Nina Hagen
by obs » Sun Sep 02, 2018 2:13 am
Scans up at Discogs already.
https://www.discogs.com/release/12459513
Nice work on making them look like the originals. Is the "Leave in Silence" jacket textured?
Yeah, the posters are cool. They should make all those promo posters available individually. I would pick up some of them, especially one for Black Celebration. Probably one from the Construction Time Again era, as well. One thing I regret was not having bought a UK or DE vinyl pressing of Construction Time. I love the black, with yellow icon. Cheapshit Canadian vinyl, of course, only had a black and white inner sleeve.
I'm surprised they didn't reprint all the 12" promo posters and included one at random .
EDIT// "Leave in Silence" IS textured. There are unboxing videos of each set on YouTube. Too lazy to find the links again, but just search for Depeche Singles unboxing and they should come up. They were uploaded by the official DM account.
by negative1 » Wed Apr 03, 2019 11:00 am
They continued with a few more boxsets in between, for Construction time again,
and Some Great Reward...
Now they are up to Black Celebration and Music for the Masses:
http://www.superdeluxeedition.com/news/ ... -12-boxes/
Depeche Mode / Black Celebration & Music for the Masses 12″ boxes
April 3, 2019 by Paul Sinclairtags: 1980s, Depeche Mode
More 12″ box sets • Seven records in 'Masses', five in 'Black Celebration'.
Depeche Mode continue with their 12-inch vinyl box set reissue campaign with packages around the singles pulled from their fifth and six albums Black Celebration and Music For The Masses.
Black Celebration was issued in March 1986 and with the band apparently bored of creating hit singles, saw a move into more gloomy atmospheric soundscapes. The three singles ('Stripped', 'A Question of Lust' & 'A Question of Time') didn't reward them with any top ten hits in Britain, the album did peak at number four in the UK charts – a new high.
Music For The Masses (it's title deliberately tongue-in-cheek) was issued in 1987 and co-produced by Dave Bascombe (best known at that point for his work with Tears For Fears). Less dense than it's predecessor, this album managed to deliver some classic singles in (particularly in 'Strangelove' and 'Never Let Me Down Again') even if, again, they weren't massive commercial successes at the time. Music For The Masses is a personal favourite for this writer – managing to be intimate, cinematic, and industrial all at the same time – and this album and the 101 tour very much laid the groundwork for the next album Violator which of course was massive, both creatively and commercially.
Black Celebration: The 12″ Singles features FIVE 'maxi singles' on vinyl; one for 'Stripped' (five tracks, 25 minutes of music), two for 'A Question of Lust' and two for 'A Question of Time.' The second 12-inch was actually a cassette-only release in the UK. This box also comes with a poster.
Music For The Masses: The 12″ Singles breaks new ground by including SEVEN 12-inch singles; two for each of the first three singles and one for slightly forgotten fourth single 'Little 15'. It too includes a poster.
Black Celebration: The 12″ Singles
A1 Stripped (Highland Mix)
A2 But Not Tonight (Extended Remix)
B1 Breathing In Fumes
B2 Fly On The Windscreen (Quiet Mix)
B3 Black Day
A Question of Lust #1
A1 A Question Of Lust
A2 Christmas Island (Extended)
B1 People Are People (Live)
B2 It Doesn't Matter Two (Instrumental)
B3 A Question Of Lust (Minimal)
A2 Christmas Island
B1 If You Want (Live)
B2 Shame (Live)
B3 Blasphemous Rumours (Live)
A Question of Time # 1
A Question Of Time (Extended Remix)
B1 Black Celebration (Live)
B2 Something To Do (Live)Mixed By – Gareth Jones
B3 Stripped (Live)Mixed By – Gareth Jones
A Question of Time #2
A Question Of Time (New Town Mix) 6:50
A2A Question Of Time (Live Remix) 4:21
B1 Black Celebration (Black Tulip Mix) 6:36
B2 More Than A Party (Live Remix)Remix 5:12
Music For The Masses: The 12″ Singles
Strangelove #1
A Strangelove (Maxi-Mix)
B1 Strangelove (Midi-Mix)
B2 Fpmip
A1 Strangelove (Blind Mix)
A2 Pimpf
B1 Strangelove (Pain Mix)
B2 Agent Orange
Never Let Me Down Again #1
A Never Let Me Down Again (Split Mix)
B1 Pleasure, Little Treasure (Glitter Mix)
B2 Never Let Me Down Again (Aggro Mix)
A Never Let Me Down Again (Tsangarides Mix)
B1 Pleasure, Little Treasure (Join Mix)
B2To Have And To Hold (Spanish Taster)
Behind The Wheel #1
A Behind The Wheel (Remixed By Shep Pettibone)
B Route 66 (Remixed By The Beatmasters)
A Behind The Wheel (Beatmasters Mix)
B Route 66 (Casualty Mix)
A Little 15
B1 Stjarna
B2 Sonata Nº 14 In C#M "Moonlight Sonata"
They did not do Shake the disease, or fly on the windscreen though.
by obs » Sun Apr 07, 2019 3:41 pm
Oh yeah, no Shake or Heart, yet they included Balance with the Construction Time set. They could have done an interim boxset, seeing as there were four 12"es. Maybe they'll do a Singles 81-85 box, double vinyl, with the four 12"es, and add the foil poster that was included with the Intercord release.
The poster for Strangelove is garbage.
And Corbijn ruined the original icons. He should stick to video work, and leave design to people who don't churn out rotten crap.
by postpunkmonk » Wed Apr 10, 2019 8:19 am
obs wrote: And Corbijn ruined the original icons. He should stick to video work, and leave design to people who don't churn out rotten crap.
For a graphic designer, Anton Corbijn sure is a great photographer/director!
https://postpunkmonk.com
For further rumination on the Fresh New Sound of Yesterday®
postpunkmonk
Location: NC | USA
by Jim2 » Thu Apr 11, 2019 7:38 am
I would like to think that there will be two additional box sets of "orphaned" 12 inch singles that would be released after they've done all the ones from their albums.
Here's what I think would be left (apologies if I missed something)
Shake The Disease (12Bong8)
Shake The Disease (L12Bong8)
It's Called A Heart (12Bong9)
It's Called A Heart (Extended) (L12Bong9)
Everything Counts, Nothing, Sacred, A Question Of Lust (12Bong16)
Remixes Of Everything Counts, Nothing And Strangelove (12Bong16)
Only When I Lose Myself (12Bong29)
Only When I Lose Myself (L12Bong29)
Enjoy The Silence···04 (12Bong34)
Enjoy The Silence···04 (L12Bong34)
Enjoy The Silence···04 (XL12Bong34)
Personal Jesus 2011 (12BONG43)
I will not buy any box sets after Exciter, although I already own far too many different copies of I Feel Loved
I Feel Loved - Mute (12BONG31)
I Feel Loved - Mute (L12BONG31)
I Feel Loved - Mute (P12BONG31) PROMO
I Feel Loved - Mute (PL12BONG31)PROMO
I Feel Loved - Reprise (9 42398-0) 2x12" Single
I Feel Loved - Reprise (PRO-A-100714-A) PROMO
by negative1 » Thu Apr 11, 2019 11:23 am
obs wrote: Oh yeah, no Shake or Heart, yet they included Balance with the Construction Time set. They could have done an interim boxset, seeing as there were four 12"es. Maybe they'll do a Singles 81-85 box, double vinyl, with the four 12"es, and add the foil poster that was included with the Intercord release.
with all the complaints about prices, it would have gotten even higher.
get the balance right was from that era. And these 2 tracks and b-sides were just for the greatest
hits compilations.
although fly on the windscreen did make it onto the album.
they can put whatever they want on these boxsets, and they will still sell. they're for collectors and completists only.
by crystal » Fri Apr 12, 2019 12:40 am
negative1 wrote: get the balance right was from that era. And these 2 tracks and b-sides were just for the greatest
Well, the 'It's called a heart' single was recorded around July 85 + they started recording 'Black celebration' in November 85. So, the time frame is pretty close + surely more of the era than 'Get the balance right' which was recorded in late 82 + was always seen and recorded as a one-off single.
Although, obviously they felt releasing a boxset with just 2 singles from that particular album, the content was a bit thin, that is why it is included. Musically it doesn't make much sense.
But totally agree with you, it doesn't really matter what they include, it will be bought anyway at almost whatever price And they can always rerelease in reshuffled form in 5 years and cash in again.
@Jim2 there was a 12" for 'Martyr' as well [Mute / 12Bong39] - though some may argue that the song is better forgotten anyway...
by Spudboy » Fri Apr 12, 2019 5:46 pm
I don't know.... I was happy when these all came out in CD box sets some 15 years ago.
Still looking for my girl with auburn hair and tawny eyes...
Spudboy
by negative1 » Fri Apr 12, 2019 6:28 pm
Spudboy wrote: I don't know.... I was happy when these all came out in CD box sets some 15 years ago.
yes, but there were a lot of promo mixes on cd and vinyl that were left off, and still aren't out yet.
some came out digitally with the other 2 remix collections though.
by ZippY » Sat Apr 13, 2019 1:33 am
Still hoping/waiting for a "box 7" of CD singles. The Exciter-period was cut-off after it's first single (Dream On) at the end of box 6.
by Jim2 » Sat Apr 13, 2019 9:16 am
crystal wrote: @Jim2 there was a 12" for 'Martyr' as well [Mute / 12Bong39] - though some may argue that the song is better forgotten anyway...
I did say "(apologies if I missed something)" and I agree, that Martyr is not memorable.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 9,944
|
El almandino o almandita es una especie de mineral perteneciente al grupo de los granates. El nombre es una corrupción de «alabandicus», el cual es el nombre aplicado por Plinio el Viejo a una piedra que se encuentra en Alabanda, una ciudad de Caria en el Asia Menor. El almandino es un granate con aluminio y hierro, que forma una serie con el piropo y otra serie con la espesartina. Es de color rojo oscuro, con tintes púrpuras. Con frecuencia se talla como roseta o cabujón. Visto a través del espectroscopio en una luz intensa, por lo general muestra tres bandas de absorción características.
Características químicas
En la serie que forma con el piropo, conocida en petrología como serie de la piralspita, los términos intermedios están entre la fórmula del almandino, Fe3Al2(SiO4)3 y la del piropo, Mg3Al2(SiO4)3, con proporciones variables entre Fe y Mg. En la serie entre piropo y espesartina, cuya fórmula es Mn3Al2(SiO4)3, los términos intermedios tienen relaciones variables entre Fe y Mn. Por supuesto, existen intermedios ternarios.
Formación y yacimientos
Es el más común de la familia de los granates. Se le encuentra frecuentemente en micaesquistos resultantes del metamorfismo regional de sedimentos arcillosos, donde es un mineral índice del grado de metamorfismo.También puede aparecer como producto metamórfico de rocas pelíticas, en rocas ácidas ígneas y como producto detrítico en sedimentos derivados de las rocas citadas.
El almandino aparece en gran cantidad en los yacimientos de Sri Lanka. También se han encontrado yacimientos en Australia, en el Tirol, este de África y Estados Unidos.
El granate almandino que aparece en el Monasterio de San Bernardo, cerca de Toledo, se conoce al menos desde el siglo XVI. Otro yacimiento español muy conocido es el «El Hoyazo de las Granatillas», en Níjar (Almería) que recibe ese nombre por la abundancia de cristales que se encuentran libres en el suelo, y que fue explotado para obtener este material para su uso como abrasivo.
Utilización
Algunos raros ejemplares pueden tallarse y se emplean como gemas. En general se utilizan como abrasivos industriales dada su enorme dureza y su fractura angular poco común.
Referencias
Enlaces externos
Almandino , uned.es.
Almandino, mindat.org.
Almandino, webmineral.com.
Manual de almandino, Mineral Data Publishing.
Nesosilicatos
Gemas
Minerales de hierro
Minerales de aluminio
Sistema cúbico
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 6,055
|
Which gas is present in LED bulb?
Which gas is used in LED?
Is there any gas in LED bulb?
Which gas is present in bulb?
What is present in LED bulb?
What are LEDs made of?
Which gas is used in CFL?
Which gas is used in halogen lamp?
Which gas is filled in halogen bulb?
What gas is in a yellow bulb?
Why gas is used in bulb?
Answer: Argon is a commonly used gas used to fill incandescent light bulbs. It increases bulb life by preventing the tungsten filaments from deteriorating too quickly. Other gases such as helium, neon, nitrogen and krypton are also used in lighting.
LED bulbs do not have any kind of gas inside them to function. Other bulbs on the market like incandescents and CFLs have gas inside and it has proved to be dangerous.
Complete step-by-step answer: Argon is a commonly used gas, used to fill incandescent light bulbs. It increases the bulb life by preventing the tungsten filament from deteriorating too quickly. Other gases such as helium, neon nitrogen and krypton are also used in lightning.
An LED light bulb is a solid-state lighting (SSL) device that fits in standard screw-in connections but uses LEDs (light-emitting diodes) to produce light. LED light bulbs are a more environmentally-friendly alternative to incandescent bulbs.
The material used most often in LEDs is gallium arsenide, though there are many variations on this basic compound, such as aluminum gallium arsenide or aluminum gallium indium phosphide.
IT IS AMAZING: What kind of lightbulb goes in a refrigerator?
In a CFL, an electric current is driven through a tube containing argon and a small amount of mercury vapor. This generates invisible ultraviolet light that excites a fluorescent coating (called phosphor) on the inside of the tube, which then emits visible light.
The early history of the halogen lamp parallels that of the incandescent. The usage of chlorine to prevent blackening of the lamp was patented in 1882. In 1959, General Electric patented a commercially viable halogen lamp using iodine as the halogen gas.
Just like with incandescent light bulbs, the electrical current enters the socket and travels up to the tungsten filament, heating up the filament to incandescence. Halogen light bulbs have tungsten filaments housed in a quartz capsule and filled with iodine and bromine gases .
Mercury vapor Light blue, intense ultraviolet In combination with phosphors used to generate many colors of light. Widely used in mercury-vapor lamps.
Sodium vapor (low pressure) Bright orange-yellow Widely used in sodium vapor lamps.
Argon or Neon gas is filled in electric bulbs. Argon is an inert gas which does not give rise to any chemical reaction and so would not affect the working of tungsten filament. These gases slow down the evaporation of tungsten filament.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 169
|
Brig. Peter Elwelu, a former 3rd Division UPDF commander in the eastern region is installed as 2nd Division commander.
Brig. Peter Elwelu, a former 3rd Division Uganda People's Defence Forces commander in the eastern region has been installed as 2nd Division commander, in the western region.
Elwelu was appointed the 2nd Division commander in June.
Elwelu returned to the 2nd Division headquarters as commander, where he previously served as the operations officer, before his transfer to the 3rd Division as commander.
During his tour of duty in the 2nd Division, he served under Brig. Patrick Kankiriho, who passed away in April.
Elwelu formally took over as commander of the western-based 2nd Division from Col. Fred Rugadya Akiiki, the deputy division commander, on Wednesday.
Rugadya has been acting as the commander since Kankiriho's death.
Maj. Gen. David Muhoozi, the commander of land forces, presided over the handover of instrument of powers at Makenke Barracks in Mbarara district.
Muhoozi urged Elwelu and his officers to cooperate with other security agents to promote security in the region.
"You need to know your officers and to work hand-in-hand with other security agencies to know the interests and threats in your area. Mind your method of work and use all instruments at your disposal to promote civil-military relations," he advised.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,748
|
\section{Introduction}
One of the most prominent critiques of superluminal travel within Einstein's general relativity (GR) is that any geometry that facilitates such travel must be largely sourced by a form of negative energy density \cite{Alcubierre1994,Everett1996,Pfenning1997,Hiscock1997,Krasnikov1998,Olum1998,VanDenBroeck1999,Millis1999,Visser2000,Loup2001,Natario2002,Gauthier2002,Lobo2003,Lobo2004,Lobo2007,Obousy2008,Finazzi2009,White2013}. Other concerns include the immense (magnitude) energy requirements to create a soliton, the difficulty associated with constructing a soliton from a nearly flat spacetime up to the superluminal phase, where the transported central observers become surrounded by a horizon, and the equal difficulties of driving the superluminal phase back the nearly flat spacetime.
There has been a recent uptick in interest regarding warp drives due to a set of papers made available in 2020 and 2021 claiming the construction of solutions that do not require sources with negative energy density \cite{Lentz2020,Bobrick2021,SantosPereira2021,Fell2021}, obeying the weak energy condition (WEC). These increasingly physical warp drives were the focus of a mini-session at the 16$^{th}$ Marcel Grossmann meeting that covered the history of warp drive research in academia, the recent positive energy warp drive research papers, and two articles regarding their reception \cite{Santiago2021a,Santiago2021b}. This article will concern only the paper written by the author \cite{Lentz2020}, summarizing its findings and discussing its standing in the literature. Mathematical notation will largely follow that of the original paper.
\section{Nat\'ario Class Spacetimes}
The class of relativistic spacetime metrics describing warp drive geometries in the literature are decomposed according to ``3+1'' Arnowitt-Deser-Misner (ADM) formalism~\cite{ADM}
\begin{equation}
ds^2 = -\left(N^2-N^i N_i \right) dt^2 - 2 N_i dx^i dt + h_{ij} dx^i dx^j,
\end{equation}
where the time coordinate $t$ stratifies spacetime into space-like hypersurfaces, the space metric components $h_{ij}$ evaluated at $t$ provide the intrinsic geometry of that hypersurface, and the similarly-evaluated shift vector components $N^i$ at $t$ provide the coordinate three-velocity of the hypersurface's normal. The time-like unit normal one-form is therefore proportional to the coordinate time element $\mathbf{n}^* = N dt$, and the unit normal vector $\mathbf{n}$ to the hypersurface has components
\begin{equation}
n^{\nu} = \left(\frac{1}{N}, \frac{N^i}{N} \right).
\end{equation}
For simplicity, we will use natural units $G=c=1$.
The majority of previous warp drive papers including Lentz 2021~\cite{Lentz2020} set the lapse function $N$ to unity and the hypersurface metric to be flat under Cartesian coordinates $h_{ij} = \delta_{ij}$. The non-flat geometry is therefore encoded in the three-component shift vector, $N_i$. The class of spacetimes described in this way have been coined as ``Nat\'ario spacetimes'' \citep{Bobrick2021,Santiago2021a,Santiago2021b}.
The projection of the Einstein equation onto the hypersurface normal gives the Hamiltonian constraint of a Nat\'ario spacetime
\begin{equation}
G^{\mu \nu} n_{\mu} n_{\nu} = 8 \pi T^{\mu \nu} n_{\mu} n_{\nu},
\end{equation}
with the projected stress-energy being referred to as the Eulerian energy density
\begin{equation}
T^{\mu \nu} n_{\mu} n_{\nu} = T^{00} = E.
\end{equation}
The geometric side of this energy constraint equation can be expressed in terms of the extrinsic curvature's trace $K = K^i_i$ and its quadratic hypersurface scalar $K^i_j K^j_i$
\begin{equation}
8 \pi E = \frac{1}{2}\left(- K^i_j K^j_i + K^2 \right).
\end{equation}
The combination of extrinsic curvatures expanded in terms of the shift vector components take the form
\begin{align}
K^2 - K^i_j K^j_i &= 2 \partial_x N_x \partial_y N_y + 2 \partial_x N_x \partial_z N_z + 2 \partial_z N_z \partial_y N_y \nonumber \\
&- \frac{1}{2} \left( \partial_x N_y + \partial_y N_x \right)^2 - \frac{1}{2} \left( \partial_x N_z + \partial_z N_x \right)^2 - \frac{1}{2} \left( \partial_z N_y + \partial_y N_z \right)^2. \label{Egeom}
\end{align}
The warp drive solution of Alcubierre~\cite{Alcubierre1994} set the precedent for WEC violation by requiring negative Eulerian energy throughout. Specifically, utilizing only a single component of the shift vector in the direction of motion, here taken to be along the positive z-axis, produces the renowned toroid of negative energy density about the soliton bubble of $N_z$, here displayed in Cartesian coordinates,
\begin{equation}
E_{\text{Alc}} = \frac{-1}{32 \pi} \left( \left( \partial_x N_z \right)^2 + \left( \partial_y N_z \right)^2 \right).
\end{equation}
The expansionless ($K = -1/2(\partial_x N_x + \partial_y N_y + \partial_z N_z) = 0$) elliptic relation of Nat\'ario 2002~\cite{Natario2002} restricted the energy form to the negative definite square of the extrinsic curvature
\begin{equation}
E_{\text{Nat}} = \frac{-1}{16 \pi} K^i_j K^j_i.
\end{equation}
Parabolic and hyperbolic relations remained to be explored at the start of 2020.
\section{Positive Energy Warp Drives}
The soliton geometry of Lentz 2021~\cite{Lentz2020} distinguishes itself from the previous literature in that it satisfies the WEC, even when moving at superluminal speeds. The WEC states that the energy of a spacetime is nowhere negative for any time-like observer. Mathematically, this means that for any time-like vector field $X$, the projection with the stress-energy tensor $T$ must be non-negative
\begin{equation}
X_{\mu} T^{\mu \nu} X_{\nu} \ge 0.
\end{equation}
The positive-energy soliton was identified through the construction of a set of rules sufficient to define geometries with everywhere non-negative energy. A brief presentation of the rules is given below.
Recall the expansion of the Hamiltonian constraint of Eqn.~\ref{Egeom}. Observe that the last three elements of the above expression are negative definite, while the first three are of indeterminant type. These first three terms provide opportunity for the Eulerian energy density function to be non-negative under particular configurations, so long as they are everywhere dominate over the first three terms. The next steps focus on such configurations.
The first rule is to reduce the 3D shift vector field to a single potential function, a real-valued function $\phi$ with spatial gradient relating the shift vector components
\begin{equation}
N_i = \partial_i \phi,
\end{equation}
satisfying a linear wave equation over the spatial coordinates
\begin{equation}
\partial_x^2 \phi +\partial_y^2 \phi -\frac{2}{v_h^2} \partial_z^2 \phi = \rho,
\end{equation}
where $ v_h/\sqrt{2}$ is the dimensionless wave front `speed' on the hypersurface, and $\rho$ is the wave equation source function, not to be confused with mass or energy density. This step provides nearly all the structure needed to find the first example positive energy drives.
Two more simplifications are used to set sufficient rules for positive Eulerian energy. The energy functional is reduced to a two-coordinate form $(z,x)$ by restricting $\rho$ and $\phi$ to be parameterized in the ($x,y$) plane by the $l_1$ norm $s= |x| + |y|$,
\begin{equation}
E = \frac{1}{16 \pi} \left(2 \partial_z^2 \phi \left(\rho + \frac{2}{v_h^2} \partial_z^2 \phi \right) - 4 \left( \partial_z \partial_x \phi \right)^2 \right), \label{Edens}
\end{equation}
which can be bounded from below by
\begin{align}
16 \pi E &\ge 2 \rho \times \partial_z^2 \phi \nonumber \\
&= \rho \times\frac{1}{ 2 v_h} \int_{-\infty}^{\infty} dx' \partial_r \rho(r,|x'| + |y|)|_{r= z-|\Delta x|/v_h}, \label{ineq}
\end{align}
where the Green's expressions for the shift vector is used in the last expression. From the lower bound expression, the last rule is formed to ensure the Eulerian energy density is everywhere non-negative: the energy function will be non-negative for configurations such that the local source density and the $z$-component source density gradient integrated along the intersecting `past' wave trajectories are of the same sign. In other words, the two factors in Expr.~\ref{ineq} must have the same sign.
It is from this sequence of rules that the Eulerian energy can be constrained to be non-negative. Demonstrating the fullness of the WEC takes several additional steps to understand the contributions of the Eulerian momentum and stress components, and is covered in detail in the original publication \cite{Lentz2020}, but are omitted here in the interest of space. The rules invoked are not strictly necessary to positive-energy warp drives. The solution space of physical warp drives is expected to be much larger and more diverse.
The shift vector of the positive-energy soliton created in Lentz 2021~\cite{Lentz2020} is given in Fig.~\ref{fig:Ns}. The soliton moves along the positive z axis at a speed set by the value of the shift vector at the origin of the co-moving coordinate coordinates in Fig.~\ref{fig:Ns}, which may be given arbitrary positive value. The transport logistics of the solitons are then similar to that of the Alcubierre solution. The solitons are constructed to contain a central region with minimal tidal forces, where proper time coincides with asymptotic coordinate time, and any Eulerian observer -- which in this case is free falling and whose velocity matches the shift vector -- within the central region would remain stationary with respect to the soliton. This is the region where a spacecraft would be placed.
\begin{figure}
\begin{center}
\includegraphics[width=\textwidth]{GREM_MG16_Proceedings_Collage.pdf}
\caption{(a) Projection of the shift vector components $N_z$ (left) and $N_x$ (right) along $(x,0,z)$. Propagation direction of the soliton is from left to right along the $z$-axis. The multi-compartment structure is a distinct departure from the single top-hat soliton found in Alcubierre 1994~\cite{Alcubierre1994} and Nat\'ario 2002~\cite{Natario2002}. Total integrated shift in each direction is 0. Note that the shift vector components are normalized with respect to the value of $N_z$ a the co-moving origin.
(b) Projection of the local energy density. The energy density is dominated by those regions containing hyperbolic source $\rho$, but also extends weakly to the boundaries of the wavefronts. The energy density is everywhere positive for Eulerian observers.
(c) Projection of the local volume expansion factor $\theta$. Positive and negative expansion factor are largely associated with negative and positive hyperbolic sources respectively. Non-zero expansion factor also exist in the spaces in-between hyperbolic sources along the hyperbolic wavefronts. Total integrated expansion factor is 0.
These plots are taken from Lentz 2021~\cite{Lentz2020}.}
\label{fig:Ns}
\end{center}
\end{figure}
The total energy requirements of the positive-energy solitons closely follow that of Pfenning \& Ford 1997~\cite{Pfenning1997} as applied to the Alcubierre solution
\begin{equation}
E_{tot} = \int E \sqrt{-g} d^3x.
\end{equation}
For solitons where the radial extent of the central region $R$ is much larger than the thickness of the energy-density laden boundary shell $w$ ($w \ll R$), the energy is estimated to be
\begin{equation}
E_{tot} \sim C v_s^2 \frac{R^2}{w}
\end{equation}
where $C$ is a form factor typically of order unity. The required energy for a positive-energy soliton with central region mean radius $R = 100$~m and average source thickness along the z-axis $w = 1$~m approaches a mass equivalent of $E_{tot} \sim (\text{few}) \times 10^{-1} M_{\odot} v_s^2$, which is of the same magnitude as the estimate of an Alcubierre solution of the same dimensions.
\section{Addressing the Literature}
The findings of Lentz 2021~\cite{Lentz2020} run against the common wisdom of the warp drive literature to that point and the proofs set forth in Olum 1999~\cite{Olum1998} and Lobo \& Crawford 2003~\cite{Lobo2003} stating that any superluminal spacetime must violate the WEC via violations of the null energy condition (NEC). The proofs are both based on an analysis of the Raychaudhuri equation for null geodesics, confined to spacetimes with only a single fastest (superluminal) causal path between two space-like 2-surfaces. The pre-conditions of these proofs are very local in nature and appear analogous to collapsing the interior of a warp drive soliton to a point in order to produce a single fastest causal path. The solitons of the early literature, such as Alcubierre 1994~\cite{Alcubierre1994} and Nat\'ario 2002~\cite{Natario2002}, have simple structures and can survive this limit. The example positive energy warp drive of Lentz 2021~\cite{Lentz2020} cannot undergo this limit without being destroyed. This drive therefore does not meet the pre-conditions of the proofs and exists outside their scope, implying that the proofs are not applicable.
Several warp drive papers have addressed the findings of Lentz 2021~\cite{Lentz2020} since an early manuscript of it was made publicly available \cite{Fell2021,Bobrick2021,Santiago2021a,Santiago2021a,SantosPereira2021b}. Of particular note, the papers of \cite{Santiago2021a, Santiago2021b} have made several assertions claiming that the solution of Lentz 2021~\cite{Lentz2020} cannot satisfy the WEC. Follow-up correspondence with the authors as well as discussion captured at the recent Marcel Grossmann meeting have demonstrated that these papers did not adequately analyze the contents of Lentz 2021~\cite{Lentz2020}. To summarize the discussions, Santiago et al.~\cite{Santiago2021a,Santiago2021b} argue that the Eulerian energy density of a soliton in a Nat\'ario class spacetime can be written as the sum of a total divergence and a negative definite term
\begin{equation}
E = \frac{1}{16 \pi} \left( \partial_i \left(N_i \partial_j N_j - N_j \partial_j N_i \right) - \frac{1}{2} \omega_i \omega_i \right),
\end{equation}
where $\omega_i = \epsilon_{ijk} \partial_k N_j$ is the shift vector vorticity. The divergence term is then argued to produce zero net energy if the warp drive is finite in size due to an application of the divergence theorem on the hypersurface where the integral's volume boundary is extended towards infinity where the divergence kernel quickly vanishes, implying that the total Eulerian energy of a Nat\'ario spacetime is non-positive. This argument does not hold in the case of Lentz 2021~\cite{Lentz2020} as the Eulerian energy density of the example positive energy soliton is smooth save for the boundaries $x=0$ and $y=0$ where stress-energy sources are only continuous, while a requirement of the divergence theorem is that the total divergence be at least first order smooth everywhere. The integral volume boundary therefore cannot be separated from the soliton and instead must be applied in a patchwork~\cite{Dray1994}, with some boundaries running adjacent to the 2-surfaces $x=0$ and $y=0$, where the divergence kernel is non-vanishing. Expansion beyond Nat\'ario class of spacetimes may smooth the geometry and sources of positive energy warp drives further.
\section{Further Challenges and Future Prospects}
There are still numerous challenges between the current state of physical warp drive research and a functioning prototype. I list here several of the more near-term challenges and give my perspective as to how research in these areas may be approached.
The most glaring challenge is the astronomical energy cost of even a modest warp drive, currently measured in solar masses where kilograms is closer to the threshold of human technology. Extreme energy savings is going to be necessary -- tens of orders of magnitude -- to bring the energy required for a warp drive down to a level that can be tested in a laboratory setting let alone be considered a viable transportation technology.
There exist numerous techniques for reducing the energy requirements of the Alcubierre solution, several of which have been very successful in reducing the (magnitude) energy requirements of the system in excess of thirty orders of magnitude \citep{VanDenBroeck1999,Loup2001,Krasnikov2003,Obousy2008,White2013}. Unfortunately, each one of these methods in their presented forms require negative energies themselves. One possible approach to uncovering significant energy savings is to modify one of these existing techniques to obey the WEC.
If the required energy can be sufficiently reduced, the next hurdle to approach is modeling the full life cycle of a physical warp drive (creation, acceleration, inertial motion, deceleration, and diffusion). Every previous publication in the field of warp drives has either assumed inertial motion (constant velocity) or has produced an accelerating/decelerating drive that violates the law of covariant conservation of stress-energy-momentum
\begin{equation}
\mathbf{\nabla} \cdot \mathbf{T} = \mathbf{0},
\end{equation}
that accompanies the Einstein equation. Deriving mechanisms for creation and acceleration is crucial to any experimental test.
The last hurdle I will mention is the full characterization of the sourcing fields, whether it be a plasma or other state of matter and energy. As stated by Matt Visser in the Q\&A of my talk, the specification of the drive geometry only is an incomplete description of the full solution. Stress-energy sources must be specified to close the system. In the hypothetical plasma of Lentz 2021~\cite{Lentz2020}, the stress-energy governing equations include the Maxwell equations for the electric and magnetic fields, the equations of motion for each species of matter, and various constituent equations governing the state of the Einstein-Maxwell-matter system. The total system is expected to be far too complex to provide analytical solutions, requiring numerical simulation as the primary means to specify each field of a soliton at any point in its life cycle.
\bibliographystyle{ws-procs961x669}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 1,562
|
<head>
<link type="type/css" href="/actual/my-style.css">
<script type="application/javascript" src="/actual/js.js"></script>
</head>
<div>body</div>
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 4,537
|
5 Venezuelan businesses that still generate profit
Published: - Dec 04, 2018
Economy Crisis Business Venezuela
Listen this article
The economic crisis that Venezuela is going through is under the focus of the international gaze. Are there businesses that thrive despite the crisis?
The answer is yes, according to Luis García, general manager of the Venezuelan American Chamber of Commerce and Industry (Venamcham), who affirms that Venezuelan companies are not disappearing, but are adapting to the crisis. Of course, such a claim depends on certain variables, such as changes in consumers and their needs. "If we talk about consumers, the Venezuelan consumer of today is different and in that sense, as a company, I have to know what product or service the Venezuelan is interested in acquiring," says García. For that reason, not all companies will succeed in the middle of the crisis, but only those belonging to industries that have constant consumers.
Leer en español: 5 negocios que aún dan dinero en Venezuela
Here are the five industries or businesses that come out in the middle of the crisis:
1. Alcohol
We must bear in mind that one of the strong points that companies have in the midst of this economic crisis is that, by decreasing imports, indigenous products have more presence and are more consumed because they are cheaper and more accessible. That is the case of rum, which today is very competitive in the Venezuelan market.
Years ago, the most widely consumed beverage in Venezuela was the imported Scotch whiskey, but now with the crisis, there are no longer so many foreign currencies and among the products that are scarce is whiskey. For this reason, the rum made in Venezuela has gained weight.
Read also: Venezuela: is it a good option not using the dollar?
"Due to the country's problems, the big importers of whiskey have seen their import limited and more rum has begun to be consumed," Jesús Alfaro, president of the Ron Producers Association, told BBC. He also affirmed that the increase in the production of rum has led to the creation of new Premium categories that have raised the status of the beverage, which was previously considered to be well below whiskey.
2.Entertainment
Even in difficult times, entertainment is an industry that will always attract, and in Venezuela, although the entertainment industry has also been affected, there is still a part that has made efforts to overcome the crisis and adapt to the needs of the consumer. In the field of theater, for example, there are groups that continue to organize events and festivals to reach citizens and distribute culture in the midst of the difficult situation. Likewise, movie theaters do not disappear either and offer special offers and a variety of options to purchase tickets.
In the midst of the crisis Venezuelans continue to worry about their academic training and for that reason, universities and institutes do not disappear, on the contrary, they offer options for people to continue training. The modalities of online education, for example, are a good option to avoid facing the transport deficits of the country, and in this way can be certified without the need to take face-to-face classes.
4. Used items
With hyperinflation, the prices of items such as clothing and household appliances are also inaccessible to Venezuelans. This, added to the need to generate additional income, has meant that the sale of second-hand items has increased.
5.Technology: Apps and Bitcoin
With the advance of the internet and smartphones, designing applications is a good option in the midst of the crisis, since they can reach different parts of the world and are paid in dollars. Likewise, the development of new technologies is also seeking the benefit of Venezuelan citizens. This is the case of Vikua, a technology company that has dedicated to developing new ways to pay the Metro, Trolleybus, and buses of Venezuela by means of cards or electronic devices.
On the other hand, the famous Bitcoin cryptocurrency has become a means of coping with the alarming devaluation of the bolivar. According to the BBC, in 2014 a bitcoin was equivalent to 40,000 bolivars, and at the beginning of last year, it was negotiated at 3.2 million. The way to obtain bitcoins is by calculating operations using special computers. This process consumes a lot of electricity, but the people who operate it do not have to pay for it because in Venezuela electricity is subsidized by the government.
According to Coindance, a portal of bitcoin performance statistics in different countries, between May and June of this year 3,354 bitcoins were traded in Venezuela. The amount that exceeds by 40% the number of bitcoins traded in Argentina, Colombia, Peru, Brazil, and Mexico as a whole.
LatinAmerican Post | María Fernanda Barinas Ortiz
Translated from "5 negocios que aún dan dinero en Venezuela"
By Lina Marcela Peñaloza
Another six months of tax benefits in Venezuela
Argentina: new project seeks to boost its economy
This is how Alberto Fernández would handle the Argentine debt
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,764
|
\section{Introduction}
The aim of this project is to adapt the main Cylindrical Algebraic Decomposition (CAD) algorithm for use with SMT-solvers, as part of the \textsf{SC}$^2$ Project which seeks to build collaborations between researchers in Symbolic Computation and Satisfiability Solving \cite{AAB+16a}. The work is influenced by the success of SAT/SMT solvers while conducting Conflict-Driven Clause Learning (CDCL) search based algorithms (see for example \cite{BHvMW09}), and their recent application to the domain of non-linear real arithmetic where CAD provides the only complete theory.
The main contributions of this report are the description of algorithms for adapting a CAD to changes in the input it is built with respect to: either incrementing or reducing the polynomial system by one constraint. We also describe results for a proof of concept implementation in the computer algebra system Maple. This shows promising experiment results with savings of up to half the calculation time, compared to a full CAD recalculation.
Other contributions are on the application side of Symbolic Computation including: an implementation of the Lazard projection and evaluation. The Lazard projection operator was proposed in 1994 \cite{lazard_1994}, but shortly after a flaw was found in its proof of correctness (see \cite{MH16} for details). However, recent work \cite{MPP17} has given an alternative proof (which necessitates changes to the lifting stage) renders this the most efficient complete CAD projection operator.
\par
\subsection{Terminology}
First, we will introduce some of the basic definitions needed to follow this article. We follow the presentation in the lecture notes by Jirstrand, \cite{jirstrand_1995}. We work over $n$-dimension real space $\mathbb{R}^n$ in which there is a variable ordering $x_1 \prec x_2 \prec \dots \prec x_n$.
\textbf{Definition \stepcounter{Count}\arabic{Count}} A \textit{decomposition} of the space \(\mathcal{X} \subset \mathbb{R}^n\) is a finite collection of disjoint regions (components) whose union is \(\mathcal{X}\).
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} A set is \textit{semi-algebraic} if it can be constructed by finitely many applications of $union$, $intersection$ and $complementation$ operations on sets of the form \(\{ x \in \mathbb{R}^{n}\ |\ \mathbf{f}(x) \geq 0 \}\) where \(\mathbf{f} \in \mathbb{R}[x_{1},\cdots,x_{n}]\).
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} A decomposition \(\mathcal{D}\) is \textit{algebraic} if each of its components \( x \in \mathcal{D}\) is a semi-algebraic set.
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} A finite partition of \(\mathcal{D}\) of \(\mathbb{R}^n\) is called a \textit{cylindrical decomposition} of \(\mathbb{R}^n\) if the projections of any two cells onto any lower dimensional coordinate space with respect to the variable ordering are either equal or disjoint.
\par
Note: any cylindrical decomposition of \(\mathbb{R}^{n}\) implies additional cylindrical decompositions of \(\mathbb{R}^{n-1}, \ldots, \mathbb{R}^{1}\).
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} If a cylindrical decomposition is also an algebraic decomposition, then it is a \textit{cylindrical algebraic decomposition} (CAD).
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} A CAD is traditionally produced \textit{sign-invariant} with respect to a set of input polynomials, which means that each polynomial has constant sign (positive, negative or zero) on each cell.
\textbf{Definition \stepcounter{Count}\arabic{Count}} The elements of a CAD are refereed to as \textit{cells}. Traditionally, each cell is equipped with: a \textit{cell index} which is a list of integers which defines the position of a cell in the decomposition (first index refers to $x_1$ etc.); and a \textit{sample point} of the cell. The cells we produce also come with a \textit{cell description}: a cylindrical formula.
For example, in a CAD \(\mathcal{C} \in \mathbb{R}^{2}[x,y]\), a cell could be defined by the triple
\(
[[1,1],[x<1,y<1],[0,0]]
\)
where: the first element \(([1,1])\) specifies the index of the cell (identifying it as the least with respect to both dimensions); the second element \(([x<1,y<1])\) gives the exact description of the cell; and finally the third element \(([0,0])\) specifies a sample point within the cell.
\par
\subsection{Example}
We give a visual example inspired by \href{url}{http://planning.cs.uiuc.edu/node296.html}. We first take a gingerbread face and remove the detail to leave an image resembling Figure \ref{fig:ging}. This is formed by four closed curves, each of which will be defined by a bi-variate polynomial equation.
\begin{figure}[H]
\centering
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=.95\linewidth]{ginger_outline}
\caption{Closed curve reduction}
\label{fig:ging}
\end{minipage}
\hfill
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=.95\linewidth]{cad.png}
\caption{A CAD of Fig \ref{fig:ging}. }
\label{fig:ging2}
\end{minipage}
\end{figure}
A CAD on this system of polynomials is visualised in Figure \ref{fig:ging2}.
Notice there are $37$ open cells (those of two dimensions). There are a further $28$ partially open (1-dimensional line segments) and $28$ closed cells (isolated points) giving $93$ CAD cells in total.
It is worth noting that in many industrial applications (especially those from SMT applications), the polynomial systems will not form aesthetically pleasing geometric shapes, and will often better resemble randomly generated systems.
\subsection{New objects of study}
We now formally define the two main objects of study in this report.
\textbf{Definition \stepcounter{Count}\arabic{Count}} An \textit{InCAD} of the space
$\mathbb{R}^n$ is an incremental cylindrical algebraic decomposition, caused by adding a polynomial $f_{new}$ into a polynomial system $\{f_1, \dots, f_m\} \in \mathbb{R}[x_1, \dots ,x_2]$, for which there was already has a CAD computed, producing a new CAD of $\{f_1, \dots, f_m, f_{new}\} \in \mathbb{R}[x_1,\dots,x_n]$.
\textbf{Definition \stepcounter{Count}\arabic{Count}} A \textit{ReCAD} of the space
$\mathbb{R}^n$ is a reduced cylindrical algebraic decomposition, caused by removing a polynomial $f_{m}$ from $\{f_1, \dots, f_m\} \in \mathbb{R}[x_1\, \dots,x_n]$, which already has the CAD computed, producing a new CAD of $\{f_1,..f_{m-1}\} \in \mathbb{R}[x_1,\dots,x_n]$.
\subsection{Report plan}
The report proceeds as follows. In Section 2 we describe Lazard's Projection scheme and in Section 3 the development and results of an incremental projection algorithm built upon this. Then in Section 4, we describe the lifting stage necessitated by Lazard projection and in Section 5 the development and results of an incremental lifting algorithm via the Lazard evaluation. We finish in Section 6 with a summary of possibilities for future work.
\newpage
\section{Projection}
\subsection{Lazard projection}
The present project built upon code from the open source \textsc{ProjectionCAD} package \cite{EWBD14} for \textsc{Maple}. This implemented the McCallum family of projection operators following \cite{McCallum1998} and needed to be adapted for the Lazard projection scheme.
First, we will introduce some notation. When we speak of projection, we usually refer to the maximal chain of projections. I.e. if a single projection eliminates one variable
\[
\pi_{n-1} : \mathbb{R}^n \rightarrow \mathbb{R}^{n-1} : (x_n,\ldots,x_1) \mapsto (x_{n-1},\ldots,x_1)
\]
then we may consider chaining these with the maximal chain one less than the total number of variables
\[
\pi^{*} : \mathbb{R}^n \rightarrow \mathbb{R}^{1} : (x_n,\ldots,x_1) \mapsto (x_1).
\]
\par
In this article we will refer to the Lazard projection operator \cite{lazard_1994}, \cite{MPP17} as \textit{ProjL} and the McCallum projection operator \textit{ProjM} \cite{McCallum1998}.
Algorithm \ref{alg:1} shown later performs the Lazard projection. It is almost identical to the counterpart McCallum projection function in the ProjectionCAD package. Both take all the discriminants and cross resultants of the input polynomials. The differences arise when taking coefficients from a polynomial in the sub-function Algorithm \ref{alg:2}. Lazard takes only the leading and trailing coefficients as opposed to all coefficients. Hence \(\textit{ProjL}(\mathcal{F}) \subset \textit{ProjM}(\mathcal{F}) \). Algorithm \ref{alg:1} iterates to perform the maximum chain or projections and produce the full set of projection polynomials.
\par
We will assume familiarity with the details of projection and lifting for CAD, so that we can focus more on the new contributions. Further details can be found in \cite{jirstrand_1995} or \cite{EWBD14} for the implementation at hand.
\textbf{Notation} We use \textbf{wrt} in place of with respect to.
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} Set \texttt{n\_elements}($Polys$) to be the number of elements in the set of polynomials $Polys$.
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} Set \texttt{degree}$(F,x)$ to be the degree of the polynomial $F$, wrt to variable x.
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} Set \texttt{RR}$(x)$ to be the function which returns a set of unique real roots of polynomial x.
\par
\textbf{Definition \stepcounter{Count}\arabic{Count}} Set \texttt{R}$(x)$ to be the function which reduces polynomial x to its square free unique factors, with constant multiples removed.
\par
\textbf{Notation} The three core algorithms implemented for the classical Lazard projection were: \texttt{ProjectionPolys}, \texttt{Projection} and \texttt{CoefficientSet}. \\
\texttt{ProjectionPolys} is effectively \(\pi^{*}\), \texttt{Projection} is \(\pi_{n_2}\), where $n_2$ is a parameter specified in the input and \texttt{CoefficientSet} which collects the coefficients required for the $n_2$'th Lazard projection.
\subsection{Worked example}
Before introducing the reader to the generalised algorithm, we will first go through, step by step, a bi-variate case of projection. We will later use this example to build upon for the incremental instance.
We begin by calculating the projection from the following polynomial system (\ref{eq:f1}):
\begin{equation}
\label{eq:f1}
\textbf{F}_1=\{\underbrace{x_1^2+x_2^2-1}_{f_1},\underbrace{x_1^3-x_2^2}_{f_2}\}.
\end{equation}
Here, the first polynomial is denoted $f_1$ and second denoted $f_2$. We first go through and calculate the univariate polynomials produced by projection and the corresponding real roots. Later, we will use this projection set to lift and build a CAD. We have left out the explicit output polynomials to save space but they can be easily calculated by most computer algebra systems.
\begin{equation}
\begin{array}{ll}
r1=\texttt{Resultant}(f_1,f_2,x_2), \ \ \ & \texttt{RR}(r1)=\{\alpha_1\} \approx \{0.7549\}\\
d1=\texttt{Discriminant}(f_1,x_2), \ \ \ &\texttt{RR}(d1)=\{0\}\\
d2=\texttt{Discriminant}(f_2,x_2), \ \ \ &\texttt{RR}(d2)=\{-1,1\}\\
l1=\texttt{Leading\ Coefficient}(f_1,x_2), \ \ \ &\texttt{RR}(l1)=\{\}\\
l2=\texttt{Leading\ Coefficient}(f_2,x_2), \ \ \ &\texttt{RR}(l2)=\{\}\\
t1=\texttt{Trailing\ Coefficient}(f_2,x_2), \ \ \ &\texttt{RR}(t1)=\{-1,1\}\\
t2=\texttt{Trailing\ Coefficient}(f_2,x_2), \ \ \ &\texttt{RR}(t2)=\{0\}\\
\end{array}
\end{equation}
So the complete set of isolated real roots is the union:
$
\{-1,0,\alpha_1 \approx 0.7549,1\} \\
$
In the presentation above we assigned an irrational root to a symbol and gave a decimal approximation, but the implementation would contain the exact root as an algebraic number.
Figures 3 and 4 plots the graphs of these functions along with the real roots isolated. We see they correspond to geometrically relevant features.
\begin{figure}[H]
\centering
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=\textwidth]{Proj1_2}
\caption
The blue curve is $f_1$ and the orange curve is $f_2$.}
\end{minipage}
\hfill
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=\textwidth]{Proj1}
\caption{Dotted lines show the projection roots.}
\end{minipage}
\end{figure}
\subsection{Incremental Lazard projection}
The most appealing approach for creating InCAD was to adapt the Lazard Projection algorithms only to calculate the new projection polynomials. To do this Algorithm \ref{alg:1} and Algorithm \ref{alg:2} had to be modified
The aim was to directly take the output from the previous call of \texttt{Projection}, together with the incremental polynomial and receive the incremented projection set. We outline the four required adjustments below, how we implemented these adjustments, as well as indicate their active state within the algorithm by highlighting the changes to the algorithms in blue in Algorithms \ref{alg:5} and \ref{alg:4} (\texttt{ProjectionAdd} and \texttt{ProjectionPolysAdd})
\texttt{ProjectionPolysAdd} is effectively \(\pi^{*}\) for the incremental case, \texttt{ProjectionAdd} is \(\pi_{n_2}\), where $n_2$ is a parameter specified in the input. Note that \texttt{CoefficientSet} itself was unchanged, only the argument passed to it was different. The new argument is now polynomials exclusively from the new polynomial set.
\subsubsection{Worked example}
In a similar fashion to Section $2.2$, we will guide the reader through a basic incremental example. We will start by incrementing the polynomial system $\textbf{F}_1$, with the polynomial forming a linear line, $f_3=x_2-x_1$. This forms the new system (\ref{eq:4}).
\begin{equation}
\label{eq:4}
\textbf{F}_2=\{\underbrace{x_1^2+x_2^2-1,x_1^3-x_2^2}_{\textbf{F}_1},\underbrace{x_2-x_1}_{f_3}\}
\end{equation}
The additional projection polynomials and their real roots are:
\begin{equation}
\begin{array}{ll}
r2=\texttt{Resultant}(f_1,f_3,x_2), \ \ \ &\texttt{RR}(r2)=\{\pm \alpha_2= \pm \frac{1}{\sqrt[]{2}} \approx \pm 0.7071\} \\% ,\alpha_2=\frac{1}{\sqrt[]{2}} \approx 0.7071 \}\\
r3=\texttt{Resultant}(f_2,f_3,x_2), \ \ \ &\texttt{RR}(r3)=\{-1,0\}\\
d3=\texttt{Discriminant}(f_3,x_2), \ \ \ & \texttt{RR}(d3)=\{\}\\
l3=\texttt{Leading\ Coefficient}(f_3,x_2), \ \ \ &\texttt{RR}(l3)=\{\}\\
t3=\texttt{Trailing\ Coefficient}(f_3,x_2), \ \ \ &\texttt{RR}(t3)=\{0\}\\
\end{array}
\end{equation}
So we have rediscovered two roots seen before ($0$ and $-1$) and two new ones ($\pm \alpha_2$). The total ordered set of real roots is now
\[
\{-1,-\alpha_2 \approx -0.7071, 0, \alpha_2 \approx 0.7071,\alpha_1 \approx 0.7549 ,1\} \\
\]
Figures 5 and 6 show that the two new roots correspond to the two new intersections of the straight line with the circle. The other new intersections happened to coincide with other already identified features.
\begin{figure}[H]
\centering
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=\textwidth]{Proj2_2}
\caption{The blue curve is $f_1$, the orange $f_2$ and the teal curve $f_3$.}
\end{minipage}
\hfill
\begin{minipage}[b]{0.48\textwidth}
\includegraphics[width=\textwidth]{Proj2}
\caption{Dotted lines show the projection roots $\in \mathcal{R}$.}
\end{minipage}
\end{figure}
\subsubsection{Algorithms}
The adjustments required to transfer to an incremental Lazard projection system are detailed as follows:
\begin{enumerate}
\item To output the full table of projection polynomials organised by the main variable and reinput this with incremental calls. This output would then be used as an input for \texttt{ProjectionPolysAdd}.
\item The process and pass the new polynomials into \texttt{ProjectionAdd}
\item To create a storage system of new polynomials which we could then apply the minimal number of operations required to obtain the correct Psetnew[i], for each index of i. We introduced a new table called Pset[3] storing the new polynomials created at each iteration of Algorithm \ref{alg:1}.
\end{enumerate}
\subsection{Incremental projection results}
Notice there were seven elements of the projection set to calculate in the original projection system \texttt{ProjL($F_1$)}. If we were to recompute the incremental system \texttt{ProjL($F_2)$} using the standard method, we would have to compute twelve elements of the projection (the original 7 plus 5 more). This would be more than double the workload of computing just those 5.
We split the testing into two experiments, one where we tested the projection algorithm on a test set of 80 examples, the other of 60 more straightforward cases. The first test set is of 80 pairs of tri-variate polynomial couples, each with four terms. The second test set contained 60 pairs of bi-variate polynomials, each with three variates. A pair consists of polynomial a and polynomial b, where b is the polynomial added incrementally. These pairs were created using the random polynomial generating function in Maple. The testing code is openly available\footnote{GitHub repository which includes analysis and worksheet - \newline $\href{url}{https://github.com/acr42/InCAD.git}$}.
An example from the tri-variate, four-term test (with chosen variable ordering $[z, y, x]$):
\begin{equation}
\label{eq:tri-var}
[-79x^2y^3+97y^2z^2+51x^2, 71x^3y^2+22xz-47y^2]
\end{equation}
An example from the bi-variate, three-term test (with variable ordering $[z, y]$):
\begin{equation}
\label{eq:bi-var}
[66z^5-40y^4+75yz^2+83z, -33y^2z^3+72y^3z+84yz^2-4z^2]
\end{equation}
The tests were controlled by a bash script rather than code within a single Maple session to avoid any caching which could bias the results.
\subsubsection{Experiment: Tri-variate}
When looking at the cases which were faster, they were on average faster by $30\%$, and $39\%$ slower in the other cases. The net speed difference had a mean of $28\%$ increase in performance with the incremental projection.
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{Projection}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.002743s & \rmfamily 0.002205s & \rmfamily \textcolor{green}{\textbf{24.39\%}} Smaller\\
\verb|Mean| & \rmfamily 0.06739s & \rmfamily 0.04809s & \rmfamily \textcolor{green}{\textbf{28.64\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.02475s & \rmfamily 0.013s & \rmfamily \textcolor{green}{\textbf{47.47\%}} Faster \\
\verb|Median| & \rmfamily 0.0625s & \rmfamily 0.035s & \rmfamily \textcolor{green}{\textbf{44.00\%}} Faster\\
\verb|Upper Quartile| & \rmfamily 0.09425s & \rmfamily 0.07525s & \rmfamily \textcolor{green}{\textbf{20.16\%}} Faster \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
\subsubsection{Experiment: Bi-variate}
When looking at the cases which were faster, they were on average faster by $55\%$, and $87\%$ slower in the other cases. The net speed difference had a mean of $16\%$ increase in performance with the incremental projection.
\par
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{Projection}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.0004660s & \rmfamily 0.0006425s & \rmfamily \textcolor{red}{\textbf{27.46\%}} Larger\\
\verb|Mean| & \rmfamily 0.03743s & \rmfamily 0.0315s & \rmfamily \textcolor{green}{\textbf{15.85\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.024s & \rmfamily 0.008s & \rmfamily \textcolor{green}{\textbf{66.66\%}} Faster \\
\verb|Median| & \rmfamily 0.0285s & \rmfamily 0.015s & \rmfamily \textcolor{green}{\textbf{47.39\%}} Faster\\
\verb|Upper Quartile| & \rmfamily 0.03675s & \rmfamily 0.05525s & \rmfamily \textcolor{red}{\textbf{50.34\%}} Slower \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
\begin{algorithm}
\caption{ProjectionPolys}
\label{alg:1}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A system of polynomials $polyset=\{f_1,\ldots,f_m\} \in \mathbb{R}[x_n,\ldots,x_1]$, $var=[x_n,\ldots,x_1]$
\STATE \textbf{Output:} A list of projection polynomials $\in \mathbb{R}[x_{n-1}, \dots, x_{1}] $
\STATE \textbf{Procedure} \texttt{Lazard ProjectionPolys}
\COMMENT{"Compute the full chain of projections"}
\STATE $dim \leftarrow$ Number of elements in $var$
\STATE $pset[0] = table() $
\STATE $pset[0] \leftarrow$ Primitive set from $polyset$, wrt variable $x_n$
\STATE $pset[0] \leftarrow$ Square free basis set from $pset[0]$, wrt variable $x_n$
\STATE $pset[0] \leftarrow$ Set of factors from $pset[0]$, wrt variable $x_n$
\STATE $cont \leftarrow$ Content Set from $polyset$, wrt variable $x_n$
\FOR{\texttt{i from 1 to dim-1}}
\STATE $out \leftarrow$ \texttt{Projection}$(pset[i-1], x_{n-i+1}, [x_{n-i},\ldots,x_1])$
\STATE $pset[i] \leftarrow (out \ \cup \ cont)$
\STATE $cont \leftarrow$ Content set from $pset[i]$, wrt $x_{n-i}$
\STATE $pset[i] \leftarrow$ Primitive set from $pset[i]$, wrt variable $x_{n-i}$
\STATE $pset[i] \leftarrow$ Square free basis set from $pset[i]$, wrt variable $x_{n-i}$
\STATE $pset[i] \leftarrow$ Set factors from $pset[i]$, wrt variable $x_{n-i}$
\ENDFOR
\STATE $ret \leftarrow pset[dim-1]$
\STATE $ret \leftarrow$ Remove constant multiples from $ret$
\STATE \textbf{return} $ret$
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{Projection}
\label{alg:2}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A set of polynomials $polyset=\{f_1,\ldots,f_m\} \in \mathbb{R}[x_n,\ldots,x_1]$, $var=x_n$ and $lvars=[x_{n-1},\ldots,x_1]$
\STATE \textbf{Output:} A set of projection polynomials $\mathbb{P}=\{p_1,\ldots,p_q\} \in \mathbb{R}[x_{n-1},\ldots,x_{1}] $
\STATE \textbf{Procedure} \texttt{Lazard Projection}
\COMMENT{"Lazard projection operator, \(\pi_{n-1}\)."}
\STATE $Polys \leftarrow$ Primitive set from $polyset$, wrt variable $x_n$
\STATE $Cont \leftarrow$ Content set from $polyset$, wrt variable $x_n$
\STATE $Polys \leftarrow$ Square free basis from $Polys$, wrt $x_n$
\STATE $Pset1 = table()$
\FOR{\texttt{i from 1 to n\_elements(Polys)}}
\STATE $Pol\leftarrow Polys[i]$
\STATE $clist \leftarrow$ Lazard coefficient set from $Pol$, wrt variable $x_n$
\STATE $temp \leftarrow$ Discriminant set from $Pol$, wrt variable $x_n$
\STATE $temp \leftarrow$ Remove constant multiples from $temp$
\STATE $Pset1[i] \leftarrow$ Concatenate together $temp$ and $clist$ \COMMENT{Need to make sure you deal with the case of temp being empty here}
\ENDFOR
\STATE $Pset2 = table()$
\FOR{\texttt{i from 1 to n\_elements(Polys)}}
\FOR{\texttt{j from i+1 to n\_elements(Polys)}}
\STATE $Pset2[i,j] \leftarrow$ Resultant of the two polynomials $Polys[i]$ \& $Polys[j]$, wrt variable $var$
\STATE $Pset2[i,j] \leftarrow $ Remove constant multiples from $Pset2[i,j]$
\ENDFOR
\ENDFOR
\STATE $Pset \leftarrow cont,Pset1,Pset2$ \COMMENT {Note that $Pset \in \mathbb{R}[x_{n-1},\ldots,x_{1}]$}
\STATE $Pset \leftarrow$ Remove constants from $Pset$
\STATE \textbf{return} $Pset$
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{ProjectionPolysAdd}
\label{alg:5}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A system of polynomials $polyset=\{f_1,..f_m\} \in \mathbb{R}[x_n,..,x_1]$, \textcolor{blue}{$prevprojset \in \mathbb{R}[x_n,\ldots,x_{1}]$, $newpolset\in \mathbb{R}[x_{n},\ldots,x_1]$} and $vars=[x_n,..,x_1]$ \textcolor{blue}{(Adjustment 1)}
\STATE \textbf{Output:} A list of projection polynomials $\in \mathbb{R}[x_{n-1},..,x_{1}] $
\STATE \textbf{Procedure} \texttt{Lazard ProjectionPolysAdd}
\COMMENT{"Compute the full chain of projections, \(\pi*\)"}
\STATE $dim \leftarrow$ Number of elements in $vars$ $+1$
\STATE $pset[0] = table()$
\STATE $pset[0] \leftarrow$ Primitive set from \textcolor{blue}{$newpolset$}, wrt variable $x_n$ \textcolor{blue}{(Adjustment 2)}
\STATE $pset[0] \leftarrow$ Square free basis set from $pset[0]$, wrt variable $x_n$
\STATE $pset[0] \leftarrow$ Set of factors from $pset[0]$, wrt variable $x_n$
\STATE $cont \leftarrow$ Set of contents of \textcolor{blue}{$newpolset$}, wrt $x_n$ \textcolor{blue}{(Adjustment 2)}
\FOR{\texttt{i from 1 to dim-1}}
\STATE $out \leftarrow$ \texttt{ProjectionAdd}$(\textcolor{blue}{prevprojset[i-1]},pset[i-1], x_{n-i+1},\ldots,x_1)$ \\ \textcolor{blue}{(Adjustment 2)}
\STATE $pset[i] \leftarrow (out \ \cup \ cont)$
\STATE $cont \leftarrow$ Content set from $pset[i]$, wrt variable x$_{n-i}$
\STATE $pset[i] \leftarrow$ Prime set from $pset[i]$, wrt variable $x_{n-i}$
\STATE $pset[i] \leftarrow$ Square free basis from set $pset[i]$, wrt variable $x_{n-i}$
\STATE $pset[i] \leftarrow$ Set factors from $pset[i]$, wrt variable $x_{n-i}$
\ENDFOR
\STATE $pset[dim-1] \leftarrow$ Remove constant multiples from $pset[[dim-1]$
\STATE $ret \leftarrow pset[[dim-1]$
\STATE \textbf{return} \textcolor{blue}{{$pset$} (Adjustment 1)}
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{ProjectionAdd}
\label{alg:4}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} Polynomial set $newpprojset=\{f_1,..f_m\} \in \mathbb{R}[x_n,\ldots,x_1]$, \textcolor{blue}{$oldprojset \in \mathbb{R}[x_{n-1},\ldots,x_{1}]$}, and variable ordering
\textcolor{blue}{(Adjustment 2)}
\STATE \textbf{Output:} Set of polynomials $Pset=\{p_1,\ldots,p_q\} \in \mathbb{R}[x_{n-1},\ldots,x_{1}] $
\STATE \textbf{Procedure} \texttt{Lazard ProjectionAdd}
\COMMENT{The incremental version of the Lazard projection operator, \(\pi_{n-1}\).}
\STATE $Polys \leftarrow$ Primitive set from \textcolor{blue}{$newpprojset$}, wrt variable $x_n$ \textcolor{blue}{(Adjustment 2)}
\STATE $Cont \leftarrow$ Content set from \textcolor{blue}{$newpprojset$}, wrt variable $x_n$ \textcolor{blue}{(Adjustment 2)}
\STATE $Polys \leftarrow$ Square free basis set from $Polys$, wrt variable $x_n$
\STATE $Pset1 = table():$
\FOR{\texttt{i from 1 to number\_elements(Polys)}}
\STATE $Pol\leftarrow Polys[i] $
\STATE $clist \leftarrow $ Lazard coefficient set from $Pol$, wrt to $x_n$
\STATE $temp \leftarrow$ Discriminant set from $Pol$, wrt $x_n$
\STATE $temp \leftarrow$ Remove constant multiples from $temp$
\STATE $Pset1[i] \leftarrow$ concat $temp$ \& $clist$
\ENDFOR
\STATE $Pset2 = table():$
\FOR{\texttt{i from 1 to n\_elements(Polys)}}
\FOR{\texttt{j from i+1 to n\_elements(Polys)}}
\STATE $Pset2[i,j] \leftarrow$ Resultant of $Polys[i]$ and $Polys[j]$, wrt to variable $var$
\STATE $Pset2[i,j] \leftarrow$ Remove constant multiples from $Pset2[i,j]$
\ENDFOR
\ENDFOR \textcolor{blue}{
\STATE $oldset \leftarrow oldprojset $
\STATE $Pset3 = table():$
\FOR{\texttt{i from 1 to n\_elements(Polys)}}
\FOR{\texttt{j from 1 to n\_elements(oldset)}}
\STATE $Pset3[i,j] \leftarrow $ Resultant of $Polys[i]$ and $oldset[j]$, wrt to variable $var$
\STATE $Pset3[i,j] \leftarrow $ Remove constant multiples from $Pset3[i,j]$
\ENDFOR
\ENDFOR \COMMENT{Adjustment 3}}
\STATE $Pset \leftarrow$ concat $(cont,Pset1,Pset2,\textcolor{blue}{Pset3})$ \COMMENT{$Pset \in \mathbb{R}[x_n,\ldots,x_{1}]$}
\STATE $Pset \leftarrow$ Remove constant multiples from $Pset$
\STATE \textbf{return} $Pset$
\end{algorithmic}
}
\end{algorithm}
\newpage
\section{Lifting}
\subsection{Lifting after Lazard projection}
We had to make changes to the lifting code not just to allow for incrementality but also changes required by the use of Lazard projection as set out in \cite{MPP17}. This requires the use of Lazard evaluation of polynomials (Algorithm \ref{alg:7}). To simplify we restricted our implementation to the open case.
\textbf{Definition \stepcounter{Count}\arabic{Count}} An \textbf{Open-CAD} is one produced by lifting over open intervals only. Thus it is not a full decomposition of ${\mathbb R}^n$ as the boundaries of the full dimensional cells are missing.
The advantage is that when building an Open-CAD we need never extend over irrational sample points, avoiding costly algebraic number calculations, but still getting a good understanding of the solution set.
We will now discuss a method used for incremental lifting, which can be thought of graphically, as a form of acyclic tree merge, later displayed visually. It will first make more sense if we explicitly show the reader the mathematical structure that allows CAD to be represented as a tree. Keeping the structure shown here in mind, will in hope make most of the algorithmic decisions, implemented during incremental lifting, much clearer.
We have two core algorithms for the Lifting. First the initial \texttt{LiftSetup} set-up algorithm (Algorithm \ref{alg:6}), and then \texttt{Lift} (Algorithm \ref{alg:8}) which iterates over the variable ordering until completion. There are also a number of other less complicated vital sub algorithms used for cell formatting/sorting/lifting purposes such \texttt{NewCadCells}, which creates new cells when lifting to a set of given roots, \texttt{GenSamplePoints} for generating sample points, \texttt{SubsFormat} which formats the sample points for substitution.
\subsection{Worked example and Lazard evaluation}
\textbf{Definition \stepcounter{Count}\arabic{Count}} Let \texttt{divide}$(x,d)$ to be the function returns boolean value true if $x$ is divisible by polynomial $d$, otherwise false.
\textbf{Definition \stepcounter{Count}\arabic{Count}} Let \texttt{SDivide}$(x,d,v)$ to be the function will completely divide polynomial $d$; which is for example $\in \mathcal{R}[x_1]$, out of polynomial $x \in \mathcal{R}[x_1,x_2]$, until the output is no longer divisible by $d$. in which case \texttt{SDivide} will then take this polynomial and substitutes the value $v$ into all instances of the variable $x_1$, returning a square free univariate polynomial in $x_2$.
We will guide the reader through a primary Lazard lift example, using Lazard evaluation (Algorithm \ref{alg:7}). The key difference to traditional CAD lifting is that we must calculate the Lazard evaluation of each polynomial at each sample point. This avoids the well-orientedness issues of the McCallum projection operators (where information is lost due to nullification at a point).
\begin{algorithm}[h]
\caption{Lazard evaluation}
\label{alg:7}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A polynomial from the projection set $f \in \mathbb{R}[x_1,\ldots,x_d]$, and $\texttt{samplepoints}=[r_1,\ldots,r_{d-1}] \in \mathbb{R}^{d-1}$.
\STATE \textbf{Output:} List of roots.
\STATE \textbf{Procedure} \texttt{Lazard\_evaluation}:
\STATE Set $\texttt{roots}$ to be an empty list
\FOR {\texttt{$j$ from $1$ to $d-1$}}
\STATE Break if cell is zero dimensional.
\FOR {\texttt{$i$ from $1$ to degree\( (f,x_j) \)}}
\IF {\texttt{f is divisible by $x_j-r_j$}}
\STATE $f \leftarrow f/(x_j-r_j)$
\ELSE
\STATE \textbf{break} innermost for loop:
\ENDIF
\ENDFOR
\ENDFOR
\STATE $f \leftarrow$ Substitute the $\texttt{samplepoints}$ into $f$
\STATE \COMMENT{$f \in \mathbb{R}[x_n,\ldots,x_d] $}
\STATE $roots \leftarrow$ Real roots of $f$
\STATE $roots \leftarrow$ Remove duplicates and sort $roots$ in ascending order
\STATE \textbf{return} roots:
\end{algorithmic}
}
\end{algorithm}
We will start by lifting the projection polynomial system defined previously \texttt{ProjL}($\textbf{F}_1$), described below. In our implementation, we only performed the lift over open intervals, as we followed the Open-CAD method. However, for this example, we will lift all cells.
\begin{equation}
\texttt{ProjL}(\textbf{F}_1)=\{x_1+1,x_1,x_1-\alpha_1 ,x_1-1\}
\end{equation}
First, we must generate our sample points for the decomposition of the real line with respect to the roots $\{-1,0,\alpha_1,1\}$.
Thus, we need a sample value from each of the following cells:
\begin{equation}
\label{ineq:1}
\begin{array}{lll}
a1=\{ x_1<-1\}, &a2=\{x_1=-1\}, &a3=\{-1<x_1<0\},\\
a4=\{x_1=0\},&a5=\{0<x_1<\alpha_1\}, &a6=\{x_1=\alpha_1\},\\
a7=\{\alpha_1<x_1<1\}, &a8=\{x_1=1\},&a9= \{1<x_1\}
\end{array}
\end{equation}
Note: $ai$ denotes the sorted $i$'th cell in the decomposition of $x_1$-space.
Our nine chosen sample points are:
\begin{equation}
\begin{array}{l}
\texttt{SamplePoints}(\texttt{ProjL}(\textbf{F}_1))=\{-2,-1,-0.5,0,0.5,\alpha_1 \approx 0.7549,0.9,1,2 \} \\
\end{array}
\end{equation}
Now we lift over each cell at the designated sample point by isolating real roots of the corresponding univariate polynomials. Below, $p_{i,j}$ denotes the polynomial acquired after applying the Lazard evaluation method to the $i$'th sample point, on the $j$'th polynomial from $F_1$
\begin{equation*}
\begin{array}{l}
\textnormal{Performing lift on the first sample point of } $$-2:$$ \\
p_{1,1}=\texttt{SDivide}(f_1,(x_1+2),-2), \ \ \ \texttt{RR}(p_{1,1})=\{\} \\
p_{1,2}=\texttt{SDivide}(f_2,(x_1+2),-2), \ \ \ \texttt{RR}(p_{1,2})=\{\} \\
\\
\textnormal{Performing lift on the next sample point } $$-1:$$ \\
p_{2,1}=\texttt{SDivide}(f_1,(x_1+1),-1), \ \ \ \texttt{RR}(p_{2,1})=\{0\} \\
p_{2,2}=\texttt{SDivide}(f_2,(x_1+1),-1), \ \ \ \texttt{RR}(p_{2,2})=\{\} \\
\\
\textnormal{Performing lift on the next sample point } $$-0.5:$$ \\
p_{3,1}=\texttt{SDivide}(f_1,(x_1+0.5),-0.5), \ \ \ \texttt{RR}(p_{3,1})=\{-\beta_0 \approx -0.8660,\beta_0 \approx 0.8660\} \\
p_{3,2}=\texttt{SDivide}(f_2,(x_1+0.5),-0.5), \ \ \ \texttt{RR}(p_{3,2})=\{\} \\
\\
\textnormal{Performing lift on the next sample point } $$0:$$ \\
p_{4,1}=\texttt{SDivide}(f_1,(x_1),0), \ \ \ \texttt{RR}(p_{4,1})=\{-1,1\} \\
p_{4,2}=\texttt{SDivide}(f_2,(x_1),0), \ \ \ \texttt{RR}(p_{4,2})=\{\} \\
\end{array}
\end{equation*}
\begin{equation*}
\begin{array}{l}
\textnormal{Performing lift on the next sample point } $$0.5:$$ \\
p_{5,1}=\texttt{SDivide}(f_1,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{5,1})=\{-\beta_1 \approx -0.8660,\beta_1 \approx 0.8660\} \\
p_{5,2}=\texttt{SDivide}(f_2,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{5,2})=\{-\beta_2 \approx -0.3536,\beta_2 \approx 0.3536\} \\
\\
\textnormal{Performing lift on the next sample point } \alpha_1: \\
p_{6,1}=\texttt{SDivide}(f_1,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{5,1})=\{-\beta_3 \approx -0.6559,\beta_3 \approx 0.6559\} \\
p_{6,2}=\texttt{SDivide}(f_2,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{5,2})=\{-\beta_4 \approx -0.6559,\beta_4 \approx 0.6559\} \\
\\
\textnormal{Performing lift on the next sample point } $$0.9:$$ \\
p_{7,1}=\texttt{SDivide}(f_1,(x_1-0.9),0.9), \ \ \ \texttt{RR}(p_{7,1})=\{-\beta_5 \approx -0.4359,\beta_5 \approx 0.4359\} \\
p_{7,2}=\texttt{SDivide}(f_2,(x_1-0.9),0.9), \ \ \ \texttt{RR}(p_{7,2})=\{-\beta_6 \approx -0.8538,\beta_6 \approx 0.8538\} \\
\\
\textnormal{Performing lift on the next sample point } $$1:$$ \\
p_{8,1}=\texttt{SDivide}(f_1,(x_1-1),1), \ \ \ \texttt{RR}(p_{8,1})=\{0\} \\
p_{8,2}=\texttt{SDivide}(f_2,(x_1-1),1), \ \ \ \texttt{RR}(p_{8,2})=\{-1,1\} \\
\\
\textnormal{Performing lift on the next sample point } $$2:$$ \\
p_{9,1}=\texttt{SDivide}(f_1,(x_1-2),2), \ \ \ \texttt{RR}(p_{9,1})=\{\} \\
p_{9,2}=\texttt{SDivide}(f_2,(x_1-2),2), \ \ \ \texttt{RR}(p_{9,2})=\{-\beta_7 \approx -2.8284,\beta_7 \approx 2.8284\}
\end{array}
\end{equation*}
We now have enough information to generate our CAD cells in ${\mathbb R}^2$. To describe our full CAD system more concisely, we will now define a new set of inequalities (\ref{ineq:inc}) we have acquired from the lift stage into $x_2$. For simplicity, we will display the CAD's created throughout this article using a graphical tree representation Fig (\ref{tree:1}). Here $bi,j$ denotes the sorted $j$'th inequality we have decomposed the $ai$'th inequality into, and so $[i,j]$ corresponds to the classical CAD cell index.
\begin{figure}[H]
\centering
\begin{minipage}[b]{\textwidth}
\includegraphics[width=\textwidth]{cad1}
\caption{CAD tree of $F_1$. The green nodes represent cells in the first dimension and red nodes represent cells in the second dimension.}
\label{tree:1}
\end{minipage}
\end{figure}
The new cell definitions are as given below.
\begin{align}
\label{ineq:inc}
&\textrm{Inequalities}(\textbf{F}_1) = \\
&\begin{array}{l}
b_{1,1}=\{x_2\}, \,
b_{2,1}=\{x_2<0\},b_{2,2}=\{x_2=0\}, \,
b_{2,3}=\{0<x_2\},\\
b_{3,1}=\{x_2<-\beta_0\},b_{3,2}=\{x_2=-\beta_0\},b_{3,3}=\{-\beta_0<x_2<\beta_0\},\\
b_{3,4}=\{x_2=\beta_0\},b_{3,5}=\{\beta_0<x_2 \}, \,
b_{4,1}=\{x_2<-1\},b_{4,2}=\{x_2=-1\},\\
b_{4,3}=\{-1<x_2<1\},b_{4,4}=\{x_2=1\},b_{4,5}=\{1<x_2\},\\
b_{5,1}=\{x_2<-\beta_1\},b_{5,2}=\{x_2=-\beta_1\},b_{5,3}=\{-\beta_1<x_2<-\beta_2\},\\
b_{5,4}=\{x_2=-\beta_2\},b_{5,5}=\{-\beta_2<x_2<\beta_2 \},b_{5,6}=\{x_2=\beta_2 \},\\
b_{5,7}=\{\beta_2<x_2<\beta_1 \},b_{5,8}=\{x_2=\beta_1 \},b_{5,9}=\{\beta_1<x_2\},\\
b_{6,1}=\{x_2<-\beta_3\},b_{6,2}=\{x_2=-\beta_3\},b_{6,3}=\{-\beta_3<x_2<-\beta_4\},\\
b_{6,4}=\{x_2=-\beta_4\},b_{6,5}=\{-\beta_4<x_2<\beta_4 \},b_{6,6}=\{x_2=\beta_4 \},\\
b_{6,7}=\{\beta_4<x_2<\beta_3 \},b_{6,8}=\{x_2=\beta_3 \},b_{6,9}=\{\beta_3<x_2\},\\
b_{7,1}=\{x_2<-\beta_6\},b_{7,2}=\{x_2=-\beta_6\},b_{7,3}=\{-\beta_6<x_2<-\beta_5\},\\
b_{7,4}=\{x_2=-\beta_5\},b_{7,5}=\{-\beta_5<x_2<\beta_5 \},b_{7,6}=\{x_2=\beta_5 \},\\
b_{7,7}=\{\beta_5<x_2<\beta_6\},b_{7,8}=\{x_2=\beta_6 \},b_{7,9}=\{\beta_6<x_2\},\\
b_{8,1}=\{x_2<-1\},b_{8,2}=\{x_2=-1\},b_{8,3}=\{-1<x_2<0\},\\
b_{8,4}=\{x_2=-0\},b_{8,5}=\{0<x_2<1 \},b_{8,6}=\{x_2=1 \},b_{8,7}=\{1<x_2\},\\
b_{9,1}=\{x_2<-\beta_7\},b_{9,2}=\{x_2=-\beta_7\},b_{9,3}=\{-\beta_7<x_2<\beta_7\},\\
b_{9,4}=\{x_2=\beta_7\},b_{9,5}=\{\beta_7<x_2
\}.
\end{array} \nonumber
\end{align}
\newpage
\subsection{Incremental Lazard lifting}
The two novel algorithms created for incremental lifting were the \newline \texttt{LiftSetupAdd} (Algorithm \ref{alg:10}); the incremental version of \texttt{LiftSetup} (Algorithm \ref{alg:6}), and \texttt{LiftAdd} (Algorithm \ref{alg:liftadd}); the incremental version of \texttt{Lift} (Algorithm \ref{alg:8}).
\par
The general concept of how we solved this stage of the problem was to think of it as solving a graph (tree) attachment/detachment problem. If you think of the old CAD as having a tree structure which we save, where nodes are cells, and branches link a cell to its parent (cell it projects onto) or child (decomposition in a cylinder above) cells. At each depth of the CAD/tree, say depth $p$, are all the cells within $\mathbb{R}^{p}$ before we lifted to $\mathbb{R}^{p+1}$. We first go through a worked example before describing the general process.
\par
We will now perform an incremental lift on the polynomial system $\textbf{F}_1$, incremented by a new polynomial $f_4=x_1^3+x_2^2$, forming the new system (\ref{eq:f3}).
\begin{equation}
\textbf{F}_3=\{\underbrace{x_1^2+x_2^2-1,x_1^3-x_2^2}_{\textbf{F}_1},\underbrace{x_1^3+x_2^2}_{f_4}\}
\label{eq:f3}
\end{equation}
The new system is symmetrical about the $y$ axis as you can see in Figure 8.
\begin{figure}[H]
\centering
\begin{minipage}[b]{0.4\textwidth}
\includegraphics[width=\textwidth]{Proj3_1}
\caption
The blue curve is $f_1$, the orange $f_2$ and the teal $f_4$.}
\end{minipage}
\hfill
\begin{minipage}[b]{0.4\textwidth}
\includegraphics[width=\textwidth]{Proj3}
\caption{Dotted lines show the projection roots.}
\end{minipage}
\end{figure}
We will skip the projection steps. However they are contained in detail within the Maple "Worked Examples" worksheet, available on GitHub\footnote{$\href{url}{https://github.com/acr42/InCAD.git}$}.
\begin{equation}
\texttt{ProjL}(\textbf{F}_3)=\ \{ \underbrace{x_1+1}_{\texttt{ProjL}(\textbf{F}_1)},\underbrace{x_1+\alpha_1}_{new},\underbrace{x_1,x_1-\alpha_1 ,x_1-1}_{\texttt{ProjL}(\textbf{F}_1)}\}
\end{equation}
We will be using calculations from Section 3.2 where we have already performed a full lift on the part of the system, bar the new projection polynomial of $x_1+\alpha_1$.
\textbf{Incremental Lift}
We now have an enlarged set of roots of univariate projection polynomials,
$\{-1,-\alpha_1,0,\alpha_1,1\}$.
The new additions below due to adding in $f_4$ are highlighted in blue.
Decomposition of the real line:
\begin{equation}
\begin{array}{l}
\label{ineq:inlift}
a1=\{ x_1<-1\}, a2=\{x_1=-1\},\textcolor{blue}{ a3=\{-1<x_1<-\alpha_1\}},\\
\textcolor{blue}{a4=\{x_1=-\alpha_1\}},\textcolor{blue}{a5=\{-\alpha_1<x_1<0\}}, a6=\{x_1=0\},
a7=\{0<x_1<\alpha_1\}, \\ a8=\{x_1=\alpha_1\},a9= \{\alpha_1<x_1<1\},a10= \{x_1=1\},a11= \{1<x_1\} \\
\end{array}
\end{equation}
Corresponding sample points:
\begin{equation}
\begin{array}{l}
\label{smple:inclft}
\texttt{SamplePoints}(ProjL(\textbf{F}_3))=\\
\{-2,-1,\textcolor{blue}{-0.9},\textcolor{blue}{-\alpha_1},-0.5,0,0.5,\alpha_1 \approx 0.7549,0.9,1,2 \} \\
\end{array}
\end{equation}
We first begin testing whether each fixed sample point has a new root, in the new projection polynomials, due to the incremental polynomial $f_4$. If it does, we will merge the new roots into the old roots list for that cell (in $x_1$). Once we have a list of new and old roots to life over, we then recompute a lift over that cell. If the previous sample point does not have any different roots to that cell, we keep its structure unchanged.
\begin{equation}
\begin{array}{l}
\textnormal{Performing lift on the first sample point of } $$-2:$$ \\
p_{1,3}=\texttt{SDivide}(f_3,(x_1+2),-2), \ \ \ \texttt{RR}(p_{1,3})= \textcolor{blue}{ \{\pm \beta_8 \approx \pm 0.8284 \} } \\
\textnormal{Performing lift on the next sample point } $$-1:$$ \\
p_{2,3}=\texttt{SDivide}(f_3,(x_1+1),-1), \ \ \ \texttt{RR}(p_{2,3})=\{-1,1\} \\
\textnormal{Performing lift on the next sample point } $$-0.5:$$ \\
p_{5,3}=\texttt{SDivide}(f_3,(x_1+0.5),-0.5), \ \ \ \texttt{RR}(p_{5,3})=\textcolor{blue}{\{\pm \beta_9 \approx \pm 0.3536\}} \\
\textnormal{Performing lift on the next sample point } $$0:$$ \\
p_{6,3}=\texttt{SDivide}(f_3,(x_1),0), \ \ \ \texttt{RR}(p_{6,3})=\{\} \\
\textnormal{Performing lift on the next sample point } $$0.5:$$ \\
p_{7,3}=\texttt{SDivide}(f_3,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{7,3})=\{\}
\\
\textnormal{Performing lift on the next sample point } \alpha_1: \\
p_{8,3}=\texttt{SDivide}(f_3,(x_1-0.5),0.5), \ \ \ \texttt{RR}(p_{8,3})=\{\} \\
\textnormal{Performing lift on the next sample point } $$0.9:$$ \\
p_{9,3}=\texttt{SDivide}(f_3,(x_1-0.9),0.9), \ \ \ \texttt{RR}(p_{9,3})=\{\} \\
\textnormal{Performing lift on the next sample point } $$1:$$ \\
p_{10,3}=\texttt{SDivide}(f_3,(x_1-1),1), \ \ \ \texttt{RR}(p_{10,3})=\{\} \\
\textnormal{Performing lift on the next sample point } $$2:$$ \\
p_{11,3}=\texttt{SDivide}(f_3,(x_1-2),2), \ \ \ \texttt{RR}(p_{11,3})=\{\}
\end{array}
\end{equation}
We now complete a full lift by considering the new sample points.
\begin{equation}
\begin{array}{l}
\textnormal{Performing lift on the next sample point } $$-0.9,$$ \\
p_{3,1}=\texttt{SDivide}(f_1,(x_1+0.9),-0.9), \ \ \ \texttt{RR}(p_{3,1})=\textcolor{blue}{\{\pm\beta_{10} \approx \pm 0.4359\} }\\
p_{3,2}=\texttt{SDivide}(f_2,(x_1+0.9),-0.9), \ \ \ \texttt{RR}(p_{3,2})=\{\} \\
p_{3,3}=\texttt{SDivide}(f_3,(x_1+0.9),-0.9), \ \ \ \texttt{RR}(p_{3,3})=\textcolor{blue}{\{\pm \beta_{11} \approx \pm 0.8538\} }\\
\textnormal{Performing lift on the next sample point } -\alpha_1, \\
p_{4,1}=\texttt{SDivide}(f_1,(x_1+\alpha_1),-\alpha_1), \ \ \ \texttt{RR}(p_{4,1})=\textcolor{blue}{\{\pm \beta_{12} \approx -0.6559\} }\\
p_{4,2}=\texttt{SDivide}(f_1,(x_1+\alpha_1),-\alpha_1), \ \ \ \texttt{RR}(p_{4,2})=\{\} \\
p_{4,3}=\texttt{SDivide}(f_1,(x_1+\alpha_1),-\alpha_1), \ \ \ \texttt{RR}(p_{4,3})=\textcolor{blue}{\{\pm \beta_{13} \approx -0.6559 \} }\\
\end{array}
\end{equation}
Note that $\beta_{12}$ is not equal to $\beta_{13}$, they differ after 10 significant figures.
The full list of cells decomposing the second dimension is now:
\begin{equation}
\begin{array}{l}
Inequalities(\textbf{F}_3)=\\
\{ \textcolor{blue}{b_{1,1}=\{x_2<-\beta_8\},b_{1,2}=\{x_2=-\beta_8\}}, b_{1,3}=\{-\beta_8<x_2<\beta_8\},\\ \textcolor{blue}{b_{1,4}=\{x_2=\beta_8\},
,b_{1,5}=\{\beta_8<x_2 \}}, \\
\textcolor{blue}{b_{2,1}=\{x_2<-1\},b_{2,2}=\{x_2=-1\},b_{2,3}=\{-1<x_2<0\}},\\
\textcolor{blue}{b_{2,4}=\{x_2=-0\},b_{2,5}=\{0<x_2<1 \},b_{2,6}=\{x_2=1 \},b_{2,7}=\{1<x_2\}},\\
\textcolor{blue}{b_{3,1}=\{x_2<-\beta_{11}\},b_{3,2}=\{x_2=-\beta_{11}\},b_{3,3}=\{-\beta_{11}<x_2<-\beta_{10}\}},\\
\textcolor{blue}{b_{3,4}=\{x_2=-\beta_{10}\},b_{3,5}=\{-\beta_{10}<x_2<\beta_{10} \},b_{3,6}=\{x_2=\beta_{10} \}},\\
\textcolor{blue}{b_{3,7}=\{\beta_{10}<x_2<\beta_{11}\},b_{3,8}=\{x_2=\beta_{11} \},b_{3,9}=\{\beta_{11}<x_2\}},\\
\textcolor{blue}{b_{4,1}=\{x_2<-\beta_{13}\},b_{4,2}=\{x_2=-\beta_{13}\},b_{4,3}=\{-\beta_{13}<x_2<-\beta_{12}\}},\\
\textcolor{blue}{b_{4,4}=\{x_2=-\beta_{12}\},b_{4,5}=\{-\beta_{12}<x_2<\beta_{12} \},b_{4,6}=\{x_2=\beta_{12} \}},\\
\textcolor{blue}{b_{4,7}=\{\beta_{12}<x_2<\beta_{13} \},b_{4,8}=\{x_2=\beta_{13} \},b_{4,9}=\{\beta_{13}<x_2\}},\\
\textcolor{blue}{b_{5,1}=\{x_2<-\beta_0\},b_{5,2}=\{x_2=-\beta_0\},b_{5,3}=\{-\beta_0<x_2<-\beta_9\}},\\
\textcolor{blue}{b_{5,4}=\{x_2=-\beta_9\},b_{5,5}=\{-\beta_9<x_2<\beta_9 \},b_{5,6}=\{x_2=\beta_9 \}},\\
\textcolor{blue}{b_{5,7}=\{\beta_9<x_2<\beta_0\},b_{5,8}=\{x_2=\beta_0 \},b_{5,9}=\{\beta_0<x_2\}},\\
b_{6,1}=\{x_2<-1\},b_{6,2}=\{x_2=-1\},\\
b_{6,3}=\{-1<x_2<1\},b_{6,4}=\{x_2=1\},b_{6,5}=\{1<x_2\},\\
b_{7,1}=\{x_2<-\beta_1\},b_{7,2}=\{x_2=-\beta_1\},b_{7,3}=\{-\beta_1<x_2<-\beta_2\},\\
b_{7,4}=\{x_2=-\beta_2\},b_{7,5}=\{-\beta_2<x_2<\beta_2 \},b_{7,6}=\{x_2=\beta_2 \},\\
b_{7,7}=\{\beta_2<x_2<\beta_1 \},b_{7,8}=\{x_2=\beta_1 \},b_{7,9}=\{\beta_1<x_2\},\\
b_{8,1}=\{x_2<-\beta_3\},b_{8,2}=\{x_2=-\beta_3\},b_{8,3}=\{-\beta_3<x_2<-\beta_4\},\\
b_{8,4}=\{x_2=-\beta_4\},b_{8,5}=\{-\beta_4<x_2<\beta_4 \},b_{8,6}=\{x_2=\beta_4 \},\\
b_{8,7}=\{\beta_4<x_2<\beta_3 \},b_{8,8}=\{x_2=\beta_3 \},b_{8,9}=\{\beta_3<x_2\},\\
b_{9,1}=\{x_2<-\beta_6\},b_{9,2}=\{x_2=-\beta_6\},b_{9,3}=\{-\beta_6<x_2<-\beta_5\},\\
b_{9,4}=\{x_2=-\beta_5\},b_{9,5}=\{-\beta_5<x_2<\beta_5 \},b_{9,6}=\{x_2=\beta_5 \},\\
b_{9,7}=\{\beta_5<x_2<\beta_6\},b_{9,8}=\{x_2=\beta_6 \},b_{9,9}=\{\beta_6<x_2\},\\
b_{10,1}=\{x_2<-1\},b_{10,2}=\{x_2=-1\},b_{10,3}=\{-1<x_2<0\},\\
b_{10,4}=\{x_2=-0\},b_{10,5}=\{0<x_2<1 \},b_{10,6}=\{x_2=1 \},b_{10,7}=\{1<x_2\},\\
b_{11,1}=\{x_2<-\beta_7\},b_{11,2}=\{x_2=-\beta_7\},b_{11,3}=\{-\beta_7<x_2<\beta_7\},\\
b_{11,4}=\{x_2=\beta_7\},b_{11,5}=\{\beta_7<x_2 \}
\}
\end{array}
\end{equation}
Figures 10-12 show the new CAD tree structure and its split into new and unchanged cells.
As you can see above, much of the CAD structure was able to be stored and reused drastically saving the solver from re-computation costs.
\begin{figure}[p]
\centering
\begin{minipage}[b]{0.9\textwidth}
\includegraphics[width=\textwidth]{cadunchanged}
\caption{CAD tree of \textbf{unchanged cells} from $F_1$ incremented by $f_4$, forming the CAD tree to $F_3$, where x1=$x_1$ and x2=$x_2$. The green nodes represent cells in the first dimension and red nodes represent cells in the second dimension.}
\end{minipage}
\end{figure}
\begin{figure}[p]
\centering
\begin{minipage}[b]{0.9\textwidth}
\includegraphics[width=\textwidth]{CAD2_new}
\caption{CAD tree \textbf{new cells} from $F_1$ incremented by $f_4$, forming the CAD tree to $F_3$, where x1=$x_1$ and x2=$x_2$. The green nodes represent cells in the first dimension and red nodes represent cells in the second dimension. Blue outlines around lines/ nodes represent new connections/ cells.}
\end{minipage}
\end{figure}
\begin{figure}[p]
\centering
\begin{minipage}[b]{0.9\textwidth}
\includegraphics[width=\textwidth]{CAD2}
\caption{CAD tree of $F_1$ incremented by $f_4$, forming the CAD tree to $F_3$, where x1=$x_1$ and x2=$x_2$. The green nodes represent cells in the first dimension and red nodes represent cells in the second dimension. Blue outlines around lines/ nodes represent new connections/ cells.}
\end{minipage}
\end{figure}
\subsection{Algorithms}
When incrementing the lift stage, we can think of it as starting at the root node of the old CAD tree and working our way down it, one depth level at a time, until we reach the leaves. In the process of working down the old tree, we will be creating subtrees, which will later be reconnected to the unchanged tree, to form the new incremented CAD tree.
\par
One tree (UnchangedCads) is a strict subset of the old trees nodes and connections, discovered through following the old structure down and Lazard evaluating each cell not marked as "new", at each depth $p$'s with new projection polynomials in $R^{p+1}$. Then, if there are new roots, we prune inspected cells children, and the cell is sent to the NewCads set for full re-computation. In the NewCads set, we will then use this cell to form a subtree, to later be reconnected with the UnchangedCads tree. Each cell in the NewCads list is a subtree, to later all be reconnected via source indices.
\par
When going through the old CAD tree structure, we have only two cases:
\par
$\textbf{CASE 1:}$ When a node has new roots, and thus new children:
\par
Each new root has been acquired from one of the new projection polynomials. We start by pruning all of the child branches in the old tree structure, by labelling them as "new", then performing a full lift onto the set of all projection polynomials from $\mathbb{R}^{k}, \ldots, \mathbb{R}^{n}$, where $k$ is the depth the new root was discovered. We separate these cells into a list called newcad. When we label a cell as new, effectively that halts tree growth in the UnchangedCads structure, so that later on we can just attach the branch extensions, gained from the incremental lift.
\par
When moving a cell into newcad, we make sure the cell has an updated source index, as its source cell would now be saved in a new tree structure with a new index.
\par
$\textbf{CASE 2:}$ When a node has no new roots,
\par
While going through the old tree, we check the source cell of each cell not labelled "new", to see if its parent was tagged as new, if so we mark it as "new". Otherwise, we label it "old", thus pass it into UnchangedCads. UnchangedCads is where we save the cost from unnecessary calculations, as cells in this tree are only Lazard evaluated at each new projection polynomial rather than all projection polynomials. If they have a fresh root, we move it over to the NewCads structure. Otherwise, we continue to attach it's child cells and their branches (source cell index).
\par
When moving cells into UnchangedCads, we make sure that the indexing of each cell does not clash with that of the indexing in NewCads at each depth level in the tree. We then merge the NewCads and UnchangedCads trees, forming the full incremented CAD tree. At each stage of the lift we merge-sort the list of cells, and because of the way be indexed them the source indices don't change.
\par
The main difficulties with implementing this method were:
\par
\begin{enumerate}
\item A cell creation memory system to trace through which cell in, say $\mathbb{R}^{3}$, were created by a cell in $\mathbb{R}^{2}$. Solved through a combination of the structure outlined below.
\item A function which could split cells in the first variable based on new roots appearing. We created the \texttt{LiftSplit} function for this,
\item We would need to store the real roots in the first variable of a CAD to search through the cells efficiently. Created a list passed through the lift functions which stores a sorted and unique list of roots in the first variable.
\item A CAD $\mathbb{R}^{1}$ and $\mathbb{R}^{1}$ sort and relabelling function to ensure the roots of the first variable are sorted in ascending order for optimised multiple applications of incremental lift. We created the function \texttt{Combine} to do this at the very end of the lifting process.
\item We would need a way of determining when a CAD cell at a higher stage has not been split at a lower stage, whether its source cell has been saved in the new tree structure and if so what the new source index is.
\item Thinking of cases of empty projection cells, how to use them to our advantage. We had to create many logical gates to work with these cases as well as creating an optimisation function \texttt{LiftAddOptimised}. This optimisation function activates if the first number of new projection polynomials sets were empty, it would copy the old tree structure down to a depth until it encounters a new root obtained from the new projection polynomials. We then begin the incremental algorithm from this boosted depth.
\end{enumerate}
We outlined the main issues above, and then we describe the steps we took to overcome them. We will introduce you to the new structure of a cad cell before moving through the details. A cell can be described now by
\begin{equation}
\textnormal{[[index],[constraints],[sample points],[source cell index],[split branch flag]]}
\end{equation}
\begin{equation}
\textnormal{e.g.} [[3],[-1<x<1,y<1],[0,0],[2],["new"]],
\end{equation}
For example, during the incremental CAD example we saw the cell (\ref{cell1}) below. This structure informs the system that cell [2] was split by a new root in $x$, or lifted onto the old and new roots in $y$.
\begin{equation}
\label{cell1}
[[2],[-1<x<1],[0],[1],["new"]].
\end{equation}
The next example cell (\ref{cell2}) tells us the new index of cell 3 in the second depth of the old cad tree, has now changed to four. So when we later move on to cell 3's children, we can adjust there source indices based on the new tree structure, to ensure the subtrees are connected correctly.
\begin{equation}
\label{cell2}
[[3],[-1<x<1,y<1],[0,0],[2],[4]].
\end{equation}
\begin{algorithm}
\caption{LiftSetup}
\label{alg:6}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A table of sets of projection polynomials $pset= \{p_1,\ldots,p_m\} \in \mathbb{R}[x_1,\ldots,x_n]$, $vars=[x_{1},\ldots,x_n]$
\STATE \textbf{Output:} A CAD $ cad_1 = \ [{cell1_1,\ldots,cell1_q}] \in \mathbb{R}[x_1] $ in the last variable $x_n$, with information ($LiftInf$) to construct the cells in the second last variable ($x_2$).
\STATE \textbf{Procedure} \texttt{Lazard LiftSetup}
\STATE $cad \leftarrow table():$
\STATE $roots \leftarrow []:$
\STATE $LiftInf_2 \leftarrow []:$ \COMMENT{where $LiftInf_i$ in the information required to lift CAD in $\mathbb{R}^{i-1}$ to a CAD in $\mathbb{R}^{i}$}
\FOR{\texttt{$i$ from $1$ to $n\_elements(pset[1])$}}
\STATE $roots \leftarrow$ Concatenate $roots$ with the real roots from $pset[1][i] \in \mathbb{R}[x_1]$
\ENDFOR
\STATE $roots \leftarrow$ Sort and remove duplicates from $roots$
\STATE $cad_1 \leftarrow$ Generate the CAD cells in the first variable $x_1$ from $roots$
\FOR{\texttt{$i$ from $1$ to $n\_elements(cad_1)$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(pset[2])$}}
\STATE $roots \leftarrow$ \texttt{Lazard\_evaluation}$(cad_1[i],pset[2][j])$
\STATE Note: $pset[2][j] \in \mathbb{R}[x_n,x_2]$
\IF{\texttt{roots<>[]}}
\STATE $LiftInf_2 \leftarrow [LiftInf_2,[[i],[roots]]]$
\STATE Where i is the cell index (note this is not the CAD index, this could be easily implemented in place, however for ease of iteration we chose to use a scalar index) and roots is a unique list of roots in $x_2$ for the cell $c_i$.
\ENDIF
\ENDFOR
\ENDFOR \COMMENT{We now have all the required information for lifting to $x_2$}
\STATE \textbf{return} \(cad_1,LiftInf_2\)
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{Lift}
\label{alg:8}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A table of sets of projection polynomials $pset= \{p_1,\ldots,p_m\} \in \mathbb{R}[x_1,\ldots,x_n]$, $vars=[x_{1},\ldots,x_n]$
\STATE \textbf{Output:} A CAD \( cad_n = \{celln_1,\ldots,celln_q\} \in \mathbb{R}^n \)
\STATE \textbf{Procedure} \texttt{Lazard Lift}
\STATE $dim \leftarrow n\_elements(vars)$
\STATE $cad_1,LiftInf_2 \leftarrow \texttt{LiftSetup}(pset,vars)$
\FOR{\texttt{$k$ from $2$ to $dim$}}
\STATE $cad_k \leftarrow []:$
\STATE $LiftInf_{k+1} \leftarrow []:$
\STATE $cad_k \leftarrow LIFT(cad_{k-1},LiftInf_k$ ) \STATE \COMMENT{Note: This lift is just the classical collins lift operation, however we are using the Lazard operators minimal polynomial set}
\IF{\texttt{k<>dim}}
\FOR{\texttt{$i$ from $1$ to $n\_elements(cad_k)$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(pset[k+1])$}}
\STATE $roots \leftarrow \texttt{Lazard\_evaluation}(cad_k[i],pset[K+1][j])$
\STATE \COMMENT{$pset[2][j] \in \mathbb{R}[x_n,x_2]$}
\IF{\texttt{roots<>[]}}
\STATE $LiftInf_{k+1} \leftarrow [LiftInf_{k+1},[[i],[roots]]]$
\STATE \COMMENT{Where i is the cell index and roots is a unique list of roots in $x_2$ for the cell $c_i$.}
\ENDIF
\ENDFOR
\ENDFOR
\ENDIF
\ENDFOR
\STATE \textbf{return} \(cad_{dim}\)
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{LiftSetupAdd}
\label{alg:10}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A sets of new projection polynomials $psetnew$, a table of sets of all projection polynomials $psetfull$, $vars$ and previous cad cells.
\STATE \textbf{Output:} A list [NewCads, OldCad, OldRoots, UnchangedCads]. NewCads in the last variable $x_n$,
OldRoots of all real roots in last variable,OldCad which contains the previous structure, UnchangedCads which is the copied and unchanged old structure.
\STATE \textbf{Procedure} \texttt{Lazard LiftSetup}
\STATE $dim \leftarrow$ Number of elements in $vars$
\STATE $cad \leftarrow table()$
\STATE $NewRoots \leftarrow []$
\STATE $NewCads \leftarrow table()$
\STATE $UnchangedCads \leftarrow table()$
\STATE $LiftInf_2 \leftarrow []$
\FOR{\texttt{$i$ from $1$ to $n\_elements(psetnew[1])$}}
\STATE $NewRoots \leftarrow [NewRoots,RealRoots(psetnew[1][i])]$
\STATE $NewRoots \leftarrow$ Sort in ascending order and remove duplicates of $NewRoots$
\ENDFOR \textcolor{blue}{
\STATE $NewCads[1],UnchangedCads[1]=\newline
\texttt{SplitCells}(OldRoots,NewRoots,OldCad)$
\COMMENT{\textbf{Adjustment 1}}
\FOR{\texttt{$i$ from $1$ to $n\_elements(oldcad[1])$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(psetnew[2])$}}
\IF{$0<>$\texttt{Lazard\_evaluation}$(oldcad[1][i],psetnew[2][j])$}
\STATE $NewCads \leftarrow $ Concatenate $oldcad[1][i]$ to $NewCads$
\ENDIF
\ENDFOR
\ENDFOR
\FOR{\texttt{$i$ from $1$ to $n\_elements(NewCads[1])$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(psetfull[2])$}}
\STATE $roots$=\texttt{Lazard\_evaluation}$(oldcad[1][i],psetnew[2][j]$
\STATE $LiftInf_2 \leftarrow [LiftInf_2,[[i],[roots]]]$
\ENDFOR
\ENDFOR
\FOR{\texttt{$i$ from $1$ to $n\_elements(oldcad[1])$}}
\STATE \textbf{If} cell $OldCad[1][i]$'s flag is not equal to "new", then concat cell to $UnchangedCads[1]$ and update index accordingly.
\ENDFOR }
\STATE $OldRoots$= Merge $OldRoots$ and $NewRoots$
\COMMENT{\textbf{Adjustment 3}}
\STATE \textbf{return} list of outputs
\end{algorithmic}
}
\end{algorithm}
\begin{algorithm}
\caption{LiftAdd}
\label{alg:liftadd}
{\fontsize{10}{15}\selectfont
\begin{algorithmic}[1]
\STATE \textbf{Input:} A sets of new projection polynomials $psetnew$, a table of sets of all projection polynomials $psetfull$, $vars$ and previous cad cells OldCad.
\STATE \textbf{Output:} Full projection set psetfull, InCAD
\STATE \textbf{Procedure} \texttt{Lazard LiftAdd}
\STATE $[NewCads, OldCad, OldRoots, UnchangedCads] \leftarrow \texttt{LiftSetupAdd}(pset,vars)$
\STATE $dim \leftarrow$ Number of elements in $vars$
\textcolor{blue}{
\FOR{\texttt{$d$ from $2$ to $dim-1$}}
\STATE $LiftInf_{d+1}\leftarrow []$
\STATE $NewCads[d] \leftarrow LIFT(NewCads_{d-1},LiftInf_d$ )
\FOR{\texttt{$i$ from $1$ to $n\_elements(OldCad[d])$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(psetnew[d+1])$}}
\IF{$0<>$\texttt{Lazard\_evaluation}$(OldCad[d][i],psetnew[d+1][j])$}
\STATE $NewCads \leftarrow $ Concatenate $oldcad[d][i]$ to $NewCads$
\ENDIF
\ENDFOR
\ENDFOR
\FOR{\texttt{$i$ from $1$ to $n\_elements(NewCads[d])$}}
\FOR{\texttt{$j$ from $1$ to $n\_elements(psetfull[d+1])$}}
\STATE $roots$=\texttt{Lazard\_evaluation}$(OldCad[d][i],psetnew[d+1][j]$
\STATE $LiftInf_{d+1} \leftarrow [LiftInf_{d+1},[[i],[roots]]]$
\ENDFOR
\ENDFOR
\FOR{\texttt{$i$ from $1$ to $n\_elements(OldCad[d])$}}
\STATE \textbf{If} cell $OldCad[d][i]$'s flag is not equal to "new", then concat cell to $UnchangedCads[d]$ and update index accordingly.
\ENDFOR
\ENDFOR
\STATE $FinalUnchangedCADS \leftarrow []$
\STATE $FinalUnchangedCADS$ Union $UnchangedCads[d][f]$ for all cells $f$ with their corresponding flags not equal to "new".
\STATE $InCAD \leftarrow FinalUnchangedCADS$ Union $NewCads[d]$ }
\STATE \textbf{return} psetfull, InCAD
\end{algorithmic}
}
\end{algorithm}
The algorithm for the Lazard Incremental Lift adjusted as follows:
\begin{enumerate}
\item First, it lifts the NewCads based on the previously obtained information. UnchangedCads
\item Then it checks to see which cells from old CAD to move into NewCads for a further lift or to copy structure over to UnchangedCads while preserving root order.
\item Gathers the necessary lift information on NewCads, then repeats.
\end{enumerate}
\subsection{Experiments: Incremental lifting}
\par
We conducted testing for the incremental lift method on the same examples used to test the projection methods earlier.
\subsubsection{Experiment: Tri-variate}
The majority of cases were faster, on average faster by $17\%$, but there were examples up to $28\%$ slower in. The net speed difference was $7\%$ increase in performance with the incremental lift.
\par
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{Lift}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.05541s & \rmfamily 0.06838s & \rmfamily \textcolor{red}{\textbf{18.96\%}} Larger\\
\verb|Mean| & \rmfamily 0.2880s & \rmfamily 0.2687s & \rmfamily \textcolor{green}{\textbf{6.707\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.1275s & \rmfamily 0.0995s & \rmfamily \textcolor{green}{\textbf{21.96\%}} Faster \\
\verb|Median| & \rmfamily 0.207s & \rmfamily 0.164s & \rmfamily \textcolor{green}{\textbf{20.77\%}} Faster\\
\verb|Upper Quartile| & \rmfamily 0.3605s & \rmfamily 0.3523s & \rmfamily \textcolor{green}{\textbf{2.29\%}} Faster \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
\subsubsection{Experiment: Bi-variate}
When looking at the cases which were faster, they were on average faster by $31\%$, and $13\%$ slower in the other case. The net speed difference was a $30\%$ increase in performance with the incremental lift.
\par
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{Lift}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.003734s & \rmfamily 0.002903s & \rmfamily \textcolor{green}{\textbf{28.63\%}} Smaller\\
\verb|Mean| & \rmfamily 0.1778s & \rmfamily 0.1240s & \rmfamily \textcolor{green}{\textbf{30.25\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.1328s & \rmfamily 0.089s & \rmfamily \textcolor{green}{\textbf{32.96\%}} Faster \\
\verb|Median| & \rmfamily 0.163s & \rmfamily 0.1255s & \rmfamily \textcolor{green}{\textbf{23.01\%}} Faster\\
\verb|Upper Quartile| & \rmfamily 0.226s & \rmfamily 0.163s & \rmfamily \textcolor{green}{\textbf{27.87\%}} Faster \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
\subsection{Experiments: Full incremental CAD}
The complete Incremental CAD system consists only of the incremental projection and incremental lift functions combined consecutively as one function.
\subsubsection{Experiment: Tri-variate}
When looking at the cases which were faster, they were on average faster by $19\%$, and $33\%$ slower in the other cases. The net speed difference was a $12\%$ increase in performance with the full incremental CAD across all cases.
\par
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{CAD}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.05880s & \rmfamily 0.08517s & \rmfamily \textcolor{red}{\textbf{30.96\%}} Larger\\
\verb|Mean| & \rmfamily 0.3296s & \rmfamily 0.2908s & \rmfamily \textcolor{green}{\textbf{11.79\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.174s & \rmfamily 0.1305s & \rmfamily \textcolor{green}{\textbf{25.0\%}} Faster \\
\verb|Median| & \rmfamily 0.24s & \rmfamily 0.182s & \rmfamily \textcolor{green}{\textbf{24.17\%}} Faster\\
\verb|Upper Quartile| & \rmfamily 0.411s & \rmfamily 0.3435s & \rmfamily \textcolor{green}{\textbf{16.43\%}} Faster \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
\subsubsection{Experiment: Bi-variate}
When looking at the cases which were faster, they were on average faster by $38\%$. The net speed difference was thus a $38\%$ increase in performance with the full incremental CAD across all cases.
\begin{center}
\begin{tabular}{@{}*4l@{}}
\toprule[1.5pt]
\multicolumn{1}{c}{\head{CAD}} &
\multicolumn{2}{c}{\head{Results}} \\
\head{} & \head{Classical} & \head{Incremental} & \head{}\\
\cmidrule(l){1-1}\cmidrule(r){2-4}
\verb|Variance| & \rmfamily 0.003896s & \rmfamily 0.002710s & \rmfamily \textcolor{green}{\textbf{43.74\%}} Smaller\\
\verb|Mean| & \rmfamily 0.2089s & \rmfamily 0.1314s & \rmfamily \textcolor{green}{ \textbf{37.12\%}} Faster \\
\verb|Lower Quartile| & \rmfamily 0.153s & \rmfamily 0.101s & \rmfamily \textcolor{green}{\textbf{33.99\%}} Faster \\
\verb|Median| & \rmfamily 0.22s & \rmfamily 0.1305s & \rmfamily \textcolor{green}{\textbf{40.68\%} }Faster\\
\verb|Upper Quartile| & \rmfamily 0.2503s & \rmfamily 0.1633s & \rmfamily \textcolor{green}{\textbf{34.77\%}} Faster \\
\bottomrule[1.5pt]
\end{tabular}
\end{center}
We think the reason for such dramatic drops in performance in more complicated cases for the lifting stage, was due to poor choices of Maple's data-structure: in particular Maple lists which are implemented as immutable types meaning our edits of them caused separate lists to be created each time.
\section{Further work}
Our implementation needs to be refactored into more appropriate data structures as discussed above.
The next step after that would be to implement a reduction CAD system (where instead of incrementally adding polynomials we can remove them). To remove a polynomial from the CAD results in finding all those projection polynomials created from purely that source polynomial as well as resultants with only the source polynomial and another polynomial and removing them and their cad cells from the system. You would then need to merge the cells which had neighbouring cells removed, possibly having to perform a full CAD on the source cells affected. The difficulty is keeping track of the fact that single projection polynomials can be computed in different ways and so a full trace would need to be maintained.
We only explored systems in which we assumed every variable in the incremental polynomial, was already represented in the system. However, situations could arise when you need another dynamic level to the system, one in which the incrementing polynomial contains at least one unseen variable $x_{n+1}$, this would require a lot of recalculations, however, would be able to benefit from storing previous calculations, as seen with this proof of concept.
\section*{Acknowledgements}
This work described was funded by the European Union's Horizon 2020 research and innovation programme under grant agreement No H2020-FETOPEN-2015-CSA 712689 (\textsf{SC}$^2$ ).
We thank Chris Brown for an informal tutorial on the Lazard evaluation and the organisers of the \textsf{SC}$^2$ 2017 Summer School at the Max Planck Institute for Informatics in Saarbr\"{u}cken, where this took place. We also thank Sam Timms, James Davenport and Stephen Forrest for useful discussions at the Summer School.
\bibliographystyle{splncs03}
|
{
"redpajama_set_name": "RedPajamaArXiv"
}
| 5,484
|
LUP Statistics
Forskarskola ska främja tvärvetenskap vid MAX IV och ESS
Publ. year
Hilner, Emelie
Department/s
MAX IV Laboratory
In LUP since
Total This Year This Month
Downloads per country
China 20 (42%)
Sweden 8 (17%)
Germany 5 (10%)
Denmark 4 (8%)
United States of America 3 (6%)
Hong Kong (China) 2 (4%)
Ukraine 1 (2%)
Netherlands 1 (2%)
Czech Republic 1 (2%)
Italy 1 (2%)
Russian Federation 1 (2%)
United Kingdom of Great Britain and Northern Ireland 1 (2%)
The download statistics shown here have been collected since the launch of LUP in October 2007 and are updated every night. Statistics are available for all records with open access fulltexts. Efforts have been made to exclude downloads by robots and track irregular download activities.
The information on downloads per country is based on the geolocation of IP addresses and may not be completely accurate. The statistics presented here may also change retroactively when the calculation process is improved to provide more accurate results.
Statistics Last Updated
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,669
|
{"url":"https:\/\/docs.marlin.pro\/books\/polkadot-gateway\/page\/manage-using-marlinctl","text":"# Manage using marlinctl\n\nThis page assumes that marlinctl is installed already. See here if you have not installed it.\n\n#### Command tree\n\nThe marlinctl command subtree for managing Polkadot gateways resides under marlinctl gateway dot. As with all marlinctl commands, the subtree can be incrementally discovered with the --help parameter.\n\n$sudo marlinctl gateway dot --help Polkadot Gateway Usage: marlinctl gateway dot [command] Available Commands: config Configurations of project set on disk create Create gateway for polkadot blockchain destroy Destroy gateway for polkadot blockchain logs Tail logs for running gateway (polkadot) instances recreate Recreate end to end gateway (polkadot) instances restart Restart services for gateway (polkadot) instances status Show status of currently running gateway (polkadot) instances versions Show available versions for use Flags: -h, --help help for dot Global Flags: --config string config file (default is$HOME\/.marlin\/ctl\/state.yaml)\n--forceful-registry-sync forceful registry sync. Do not use if you don't know what this is for.\n--loglevel string marlinctl loglevel (default is INFO) (default \"info\")\n--skip-registry-sync skip registry sync during run\n--skip-update-check skip update check during run\n\n\n\n#### Create a gateway\n\nGateways can be created using the\u00a0create command.\n\n$sudo marlinctl gateway dot create --help Create gateway for polkadot blockchain Usage: marlinctl gateway dot create [flags] Flags: -b, --bootstrap-addr string Bridge bootstrap address -a, --chain-identity string Gateway's keystore path (default \"gateway_dot.key\") -c, --contracts string mainnet\/kovan (default \"mainnet\") -d, --discovery-addr string Bridge discovery address (default \"0.0.0.0:20702\") -h, --help help for create -i, --instance-id string instance-id of spawned up resource (default \"001\") -l, --internal-listen-address string Bridge listen address (default \"127.0.0.1:20901\") -y, --keystore-pass-path string Keystore pass path -k, --keystore-path string Keystore Path -g, --listen-addr string Address on which gateway listens for connections from peer (default \"\/ip4\/0.0.0.0\/tcp\/20900\") -p, --pubsub-addr string Bridge pubsub address (default \"0.0.0.0:20700\") -r, --runtime-args stringToString runtime arguments while starting up (default []) -s, --skip-checksum skip checksum verification while starting up binaries -x, --version string runtime version override Global Flags: --config string config file (default is$HOME\/.marlin\/ctl\/state.yaml)\n--forceful-registry-sync forceful registry sync. Do not use if you don't know what this is for.\n--loglevel string marlinctl loglevel (default is INFO) (default \"info\")\n--skip-registry-sync skip registry sync during run\n--skip-update-check skip update check during run\n##### Primary flags\n\nWhile the following flags are optional for testing the connection between the gateway and the polkadot node, they are required parameters for interacting with the Marlin network.\n\n\u2022 --bootstrap-addr - Address of the bootstrap server used for discovering other nodes in the network.\n\u2022 --keystore-path - Path to keystore file. Serves as the identity of the gateway.\n\u2022 --keystore-pass-path - Path to text file containing passphrase that can unlock the above keystore file.\n##### Secondary flags\n\u2022 --contracts - Contract set used by the gateway for fetching on-chain data. Enables choosing the network that the gateway is a part of (e.g. separate networks for testing or new upgrades). Especially important in the initial phases of the network, where it's expected that clusters will work on a testnet first before transitioning to mainnet upon maturity.\n\u2022 --chain-identity - Path to a polkadot key file which the gateway uses as its identity. Useful to give the gateway a deterministic identity on the polkadot side so deployments can be automated.\n\nThe following flags are useful for customizing the socket addresses used by the gateway in order to prevent any conflicts with other programs on the same system.\n\n\u2022 --listen-addr - Address where the gateway listens for incoming connections from your polkadot nodes.\n\u2022 --discovery-addr - Address used for the discovery protocol to find peers in the Marlin network.\n\u2022 --pubsub-addr - Address used for the pubsub protocol, to send and receive message from the Marlin network.\n\u2022 --internal-listen-addr - Address used for internal communication between components of the gateway.\n\n#### See logs\n\nLogs can be tailed using the\u00a0logs command.\n\n$sudo marlinctl gateway dot logs --help Tail logs for running gateway (polkadot) instances Usage: marlinctl gateway dot logs [flags] Flags: -h, --help help for logs -i, --instance-id string instance-id of resource to log (default \"001\") -n, --last int number of last lines to tail in logfile (default 100) Global Flags: --config string config file (default is$HOME\/.marlin\/ctl\/state.yaml)\n--forceful-registry-sync forceful registry sync. Do not use if you don't know what this is for.\n--loglevel string marlinctl loglevel (default is INFO) (default \"info\")\n--skip-registry-sync skip registry sync during run\n--skip-update-check skip update check during run\n##### Primary flags\n\u2022 --last - Number of lines from the end to tail\n\n#### Destroy a gateway\n\nGateways can be destroyed using the\u00a0destroy command.\n\n$sudo marlinctl gateway dot destroy --help Destroy gateway for polkadot blockchain Usage: marlinctl gateway dot destroy [flags] Flags: -h, --help help for destroy -i, --instance-id string instance-id of resource to destroy (default \"001\") Global Flags: --config string config file (default is$HOME\/.marlin\/ctl\/state.yaml)\n--forceful-registry-sync forceful registry sync. Do not use if you don't know what this is for.\n--loglevel string marlinctl loglevel (default is INFO) (default \"info\")\n--skip-registry-sync skip registry sync during run\n--skip-update-check skip update check during run","date":"2022-09-26 09:37:18","metadata":"{\"extraction_info\": {\"found_math\": true, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 1, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 1, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.38099509477615356, \"perplexity\": 12354.647497900874}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 20, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2022-40\/segments\/1664030334855.91\/warc\/CC-MAIN-20220926082131-20220926112131-00676.warc.gz\"}"}
| null | null |
Man who impregnated girlfriend's underage daughter gets jail and caning
File photo of the State Courts in Singapore. (Photo: Calvin Oh)
Lydia Lam
@LydiaLamCNA
29 Nov 2021 03:31PM (Updated: 29 Nov 2021 03:31PM)
SINGAPORE: A man who began having sex with his girlfriend's 14-year-old daughter while she was at work, later impregnating the minor, was sentenced to 12 years' jail and six strokes of the cane on Monday (Nov 29).
The prosecution said the 29-year-old man, who cannot be named to protect the victim's identity, had fathered the victim's child in a "disturbing twist of events" even though he had become the father figure she never had.
The offender engaged in what the prosecution called "a clandestine sexual relationship with the victim" from late 2018 to June 2020.
The prosecutors had sought between 10 and 12 years' jail and six strokes of the cane, saying that the man's offences have "irreversibly altered the trajectory of the victim's life, forcing her to assume the role and responsibilities of a mother when she is little more than a child herself".
The defence argued that the accused was "in a romantic and caring relationship with the victim" and that this was not a situation where he "went into his specific role as a parental figure" to abuse his position.
The offender pleaded guilty last month to three counts of sexual penetration of a minor, with another 16 charges taken into consideration.
He began dating the victim's mother in 2015 and later moved in to stay with her. The victim initially was not close to him, as she had been sexually abused by a family member when she was 10 to 12.
However, the offender earned the girl's trust by assuming the role of a father figure in her life, doting on her and taking her out for her favourite activities such as ice skating.
He cooked, cleaned and cared for her when she was ill and guided her with her homework, with the victim eventually calling him "daddy".
However, their relationship "developed into a sexual one" in October 2018, and the pair began having "consensual unprotected sexual intercourse regularly", the prosecutor said.
They did so until December 2019, switching to other sex acts between January and June 2020 when the victim's "pregnancy was causing too much discomfort".
The victim was admitted to hospital for labour contractions and gave birth to a baby girl in May 2020. The hospital alerted the police to the teenage pregnancy, and the Immigration and Checkpoints Authority also flagged the matter to the police a few weeks later after the victim tried to register the birth of her child and was found to be under 16.
Even after the birth of the baby, the offender continued to engage in sex acts with the victim and was arrested only in early July last year.
The prosecution said that sexual penetration of a minor in the context of an exploitative relationship is "reprehensible and must be met with the full force of the law".
"The accused lived in the home that the victim shared with her mother and was the only paternal figure she had ever known," they said.
"Not only did the accused exploit the victim's immaturity and the trust she reposed in him, he also betrayed the trust of the victim's mother, with whom he was in a romantic relationship."
The prosecutors said it was "perverse - and indeed, aggravating" that the offender engaged in sexual activity with the victim when her mother was at work as the sole breadwinner of the family.
A Child Guidance Clinic report showed that the victim felt sad that she missed out on her dream of graduating with her friend in 2020 and has "passive suicidal wishes" because of the repeated sexual abuse she experienced.
She also worries that her child might hate her if she finds out about her illegitimate birth, and suffers moodiness, irritability and difficulty sleeping.
Source: CNA/ll(rw)
court crime
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 1,472
|
@implementation UIViewController (SpaceShuttle)
- (void)presentViewController:(UIViewController *)viewControllerToPresent animated:(BOOL)flag completion:(void (^)(void))completion supplies:(NSDictionary *)supplies {
viewControllerToPresent.from = self;
[self setupViewController:viewControllerToPresent withSupplies:supplies];
[self presentViewController:viewControllerToPresent animated:YES completion:completion];
}
- (void)dismissViewControllerAnimated:(BOOL)flag completion:(void (^)(void))completion supplies:(NSDictionary *)supplies {
[self setupViewController:self.from withSupplies:supplies];
[self dismissViewControllerAnimated:flag completion:completion];
}
@end
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 3,172
|
Vidya Balan named Samajwadi Pension Scheme's brand ambassador
LUCKNOW - Bollywood actress Vidya Balan was on Friday named the brand ambassador for the ambitious 'Samajwadi Pension Scheme' of the Uttar Pradesh government. Thanking Chief Minister Akhilesh Yadav for the same, she said she was impressed by the work for women that was being done in the state and complimented him for the pension scheme, which she said would go a long way in empowering women...
and changing their lives for good. Vidya also assured of her all possible support towards value adding to the scheme. Akhilesh Yadav also announced at the event that in future, all eligible poor women in the state would be covered under the scheme, under whichRs 500 was being given per month to beneficiaries through direct bank transfer, and said this initiative had enhanced the respect and honour of women. He added that a scheme of such a mammoth scale was not being run anywhere in the country.
The chief minister also informed that under the Samajwadi Pension Scheme, the family's woman head is given priority over others and that 55 lakh poor families in the state were being covered under it.
Stressing that the scheme proves his government's commitment and concerns for the poor, helpless and the needy, he also expressed hope that it would help in improving the economic, social, educational and healthcare indices and women would take full benefit of it.
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 5,331
|
In order to make this Site as customized as possible to your needs and interests, we will gather certain information from you. We will use this information to tailor our services, content and advertising to you. In order to accomplish this we will ask you to register with the Site and provide certain personal information prior to allowing you to enter any contests, post any comments or otherwise visit and utilize certain areas of the Site. If you opt to not register then you may not have access to the areas of the Site which require your personal information.
Due to the business-oriented nature of our business and the Site, the Site is not designed to appeal to anyone under the age of 18. We do not knowingly request or receive any information from children.
We will offer a free electronic newsletter to users. The Company gathers the email addresses of users who voluntarily subscribe. Users may remove themselves from this mailing list by following the link provided in every newsletter that points users to the subscription management page. Users can also subscribe to the newsletters at the time of registration.
Our site users can choose to electronically forward a link, page, or documents to someone else by clicking "tell a friend." The user must provide their email address as well as that of the recipient. This information is used only in the case of transmission errors and, of course, to let the recipient know who sent the email. The information is not used for any other purpose.
We may offer interactive polls to users so they can easily share their opinions with other users and see what our audience thinks about important issues. Opinions or other responses to polls are aggregated and are not identifiable to any particular user. The Company may use a system to "tag" users after they have voted, only so that they can vote just once on a particular question. This tag is not correlated with information about individual users.
The Company may occasionally conduct user surveys to improve our products and our customer service. We never share any of this information about specific individuals with any third party.
Provided Information: The personal information which you provide to us, including your name, address and email address, while using the Site, such as during the account registration process, responding to surveys, uploading videos, posting messages or comments, entering contests, subscribing to newsletters or other mailing lists or entering promotions.
Web Beacons: While you are on the Site, we may use Web Beacons or clear GIFs. These tools allow use to track the online usage patterns of our customers in a non-personally-identifiable fashion. We may also use Web Beacons in HTML-based emails sent to you by the Company or its affiliates in order to determine whether or not the emails have been read.
Placing an Order: To place online orders and check order status via the Site, you must create a profile. This information is essential to use our services correctly. Mandatory fields include: username and password you create, e-mail address and mailing address. We may request authorization to contact you for specials and promotional offers, but it is not mandatory. The information you give is not transferred, shared or sold, unless you authorize consent or we are legally required. However, an exception to sharing information is when it is needed by a Third Party that helps us serve you, such as providing your name and address to a shipping agent or payment processor.
Log File Information: Whenever you visit the Site, our web servers will automatically record certain information from the web server from which you are using, including the IP address, the type of browser which you are utilizing, the pages viewed from the web browser and other usage details available from your server. This information will allow us to assist in the analysis and correction of technical problems and to better protect the equipment of our users and ourselves.
Applications you use or download from our Site may contain digital rights management systems ("DRMS") from the Company or third parties. DRMS provide a means of communication between your software and the third parties and allows third parties to embed certain security features into the software. DRMS are often used to prevent the distribution of or access to software in the event of unauthorized use. These DRMS are subject to their own license agreements. By your use of this Site, you agree that the Company shall not be responsible for any loss or damage of any sort relating to the use of the DRMS or your dealings with such third parties.
We use industry-standard security measures in place to protect the loss, misuse, and alteration of the information under our control, and we make good-faith efforts to store your information in a secure operating environment. However, no data transmission over the Internet can be guaranteed to be 100% secure. As a result, while we strive to protect your information, we cannot ensure or warrant the security of any information you transmit to us, and you do so at your own risk. If you created an account on the Site with a User ID and password to access portions or features of the Site, the Company disclaims any responsibility for a breach of privacy as a result of your willingly or inadvertently disclosing your User ID and/or password to others.
Sensitive information is always stored under a password, thus adding an additional layer of security. When you set up an account at myheritagecards.com, you choose your own password for the account and you control who has access to your account. You may change your password at any time.
Upon placing an order, myheritagecards.com will send you an email confirming the order. This will not only inform your order's progress, but will also alert you if someone is using your account or computer without your permission. If you receive a confirmation for an order you did not place, please email us immediately at: billing@myheritagecards.com.
If you would like to change your personally identifiable information in our databases, you may do so through your Account Profile when you are "logged in" to the Site, or by contacting support@myheritagecards.com. We will make all reasonable efforts to comply with your request. Keep in mind, however, that there will be residual information that will remain within the Company databases, access logs, and other records, which may or may not contain inaccurate or outdated personally identifiable information. The residual information will not be used for commercial purposes; however, the Company reserves the right, from time to time, to re-contact former users of our Services for the purpose of informing them of changes in our service or hours of service.
The Company customers are asked to provide their email address upon ordering cards from us. This is used to send notification of updates and changes in our services and/or products, as well as to communicate with our customers about their orders.
The Company uses a few third parties (FirstData(Clover), Paypal, Stripe, Square) for credit card processing. These parties are solely a link in the distribution chain, and are not permitted to store, retain, or use the information provided except for the sole purpose of credit card processing.
We do not disclose to any third party the information provided by our customers. All financial/billing information is used solely to invoice our customers for our services and products. This information is not used for marketing purposes.
Notwithstanding anything else to the contrary, the Company is free to disclose your personal information if it is required to do so by law or if we have a good faith belief that disclosure is necessary to (1) comply with the law or with legal process served on us; (2) protect and defend our rights or property; or (3) act in an emergency to protect someone's safety.
With permission, the Company reserves the right to utilize or reproduce the name, photo and/or information contained on trading cards, or trading cards themselves, as samples, for promotional/advertising purposes including as part of our site gallery.
After you create an account on the Site and "opt in" to providing your personal information, you may at any time unsubscribe from the Site. In order to "opt out," please email us at privacy@myheritagecards.com. Please note that there might be a delay between when you submit your email to opt-out and when it is processed. We appreciate your patience during this period. We maintain archives of our web logs, database, and other systems and information for legal, disaster recovery and other purposes. If you request that we delete your information from our active database, please understand that it is possible some of your information may remain archived after we delete the information from its active database.
The Company does not knowingly collect or maintain the personally identifiable information of children under the age of thirteen. The Site is not directed to children under the age of thirteen. If you are under the age of thirteen, we ask that you do not use or view the Site. If the Company learns that it has collected information from someone under the age of thirteen, then the Company will take all steps necessary to delete such personal information. If you believe that your child has provided his or her information to us, please send us an email or a written letter at the address below so that we may delete such information.
If the Company is acquired by or merged with another entity, the Company shall have the right to transfer or assign the information we have collected from you in connection with such corporate transaction. In the unlikely event of our bankruptcy, insolvency, reorganization, receivership, or assignment for the benefit of creditors, or the application of laws or equitable principles affecting creditors' rights generally, we may not be able to control how your personal information is treated, transferred, or used and assume no liability for any such improper distribution or use.
Please visit our Terms & Conditions section establishing the use, disclaimers, and limitations of liability governing the use of our Site at: www.myheritagecards.com/TermsConditions.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 7,341
|
Дом купцов Степановых находится в Перми на улице Ленина. Входит в число памятников архитектуры.
История
История дома начинается в 1873 г., когда 25 октября 1873 г. пермский купец Дмитрий Степанович Степанов подал в Городскую думу проект строительства полукаменного двухэтажного дома рядом с деревянным домом, возведённым им в 1856 г. вместе со служебными постройками и использовавшимся для торговли табачными изделиями, колониальными товарами и бакалеей. Этот проект был одобрен уже на следующий день, но 22 апреля 1874 г. он попросил отозвать прежний проект и подал новый — на строительство каменного двухэтажного дома. Архитектор Р. И. Карвовский выдал справку о возможности строительства дома на Покровской улице по поданному проекту. 25 апреля разрешение на постройку дома было выдано.
Купец Дмитрий Степанов сдавал этот дом в аренду и имел с него неплохую прибыль. Его сын Александр Дмитриевич Степанов продолжил дело отца. В 1885 г. в «Пермских губернских ведомостях» сообщалось, что в доме открыт розничный магазин мануфактуры «Жаккардовой фабрики Гилле и Дитрих». В 1998 г. дом принадлежал купцу Дмитрию Степановичу Степанову и оценивался в 8 тысяч рублей. Сохранилась реклама 1904 г., что в магазине по Покровской улице продаётся широкий набор табачных, кондитерских изделий.
В 1909 г. А. Д. Степанов начал сдавать второй этаж дома для организованной в городе телефонной станции Пермского почтово-телеграфного округа на 625 абонентов. В 1912 г. располагалась типография почётного гражданина Перми Чердынцев Владимира Алексеевича. В 1916 г. дом оценивался в 13380 руб. и упоминается как принадлежащий купчихе Полаженко Любови Дмитриевне.
В 1918 году на основании Декрета ВЦИК от 20 августа 1918 г. был муниципализирован, а в 1924 г. домовладение было описано как находящееся в ведении Коммунхоза и сдаваемое в аренду. В 1928 г. дом прошёл техническое освидетельствование на предмет возможности достройки к нему третьего этажа, но это не было реализовано.
В 1990-е гг. дом Степановых находился на балансе управления связи, а его помещения занимал детский сада № 24. Дом был принят под охрану государства в соответствии с решением Малого Совета Пермского Облсовета № 683 20 мая 1993 г. как памятник архитектуры «Жилой дом Н. П. Падалка». В дальнейшем памятник архитектуры был переименован в «Дом жилой Д. С. Степанова» в государственных списках памятников истории и культуры Пермской области, утверждённых распоряжением губернатора области 5 декабря 2000 г. № 713-р.
Примечания
Ссылки
Литература
Культура Перми
Степановых
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 2,407
|
{"url":"http:\/\/sites.millersville.edu\/bikenaga\/basic-algebra\/radical-equations\/radical-equations.html","text":"In this section, I'll discuss how you solve equations involving square roots of variable expressions.\n\nThe idea in solving such equations is to square both sides of the equation--- sometimes several times --- to eliminate the radicals. It's important to check the solutions when you're done, because it's possible for this procedure to produce bogus solutions.\n\nExample. Solve .\n\nSquare both sides:\n\nThen\n\nCheck: When ,\n\nThe solution is .\n\nExample. Solve .\n\ncan't be negative, because denotes the nonnegative square root by definition. Therefore, the equation has no solutions.\n\nExample. Solve .\n\nSquare both sides:\n\nNote that in multiplying out I was careful to remember the middle term (\" \"). Forgetting the middle term is one of the most common mistakes made in solving this kind of problem.\n\nNow I have\n\nFactor and solve:\n\nchecks when substituted in the original equation. However, gives\n\nSo the only solution is .\n\nExample. Solve .\n\nSquare both sides and multiply out:\n\nNote that in the second step I was careful to remember the middle term in computing .\n\nBefore squaring again, I want to isolate the square root. Otherwise, I'll just create more square root terms and I won't make any progress.\n\nSo\n\nNow square both sides and multiply out:\n\nThen\n\nCheck: If ,\n\nThe solution is .\n\nExample. Solve .\n\nSquare both sides:\n\nNow square both sides again:\n\nSolve for x:\n\nCheck: When ,\n\nWhen ,\n\nis not a solution.\n\nThe only solution is .\n\nExample. Solve .\n\nSquare both sides:","date":"2019-03-26 03:59:10","metadata":"{\"extraction_info\": {\"found_math\": false, \"script_math_tex\": 0, \"script_math_asciimath\": 0, \"math_annotations\": 0, \"math_alttext\": 0, \"mathml\": 0, \"mathjax_tag\": 0, \"mathjax_inline_tex\": 0, \"mathjax_display_tex\": 0, \"mathjax_asciimath\": 0, \"img_math\": 0, \"codecogs_latex\": 0, \"wp_latex\": 0, \"mimetex.cgi\": 0, \"\/images\/math\/codecogs\": 0, \"mathtex.cgi\": 0, \"katex\": 0, \"math-container\": 0, \"wp-katex-eq\": 0, \"align\": 0, \"equation\": 0, \"x-ck12\": 0, \"texerror\": 0, \"math_score\": 0.9279583096504211, \"perplexity\": 1273.549383713934}, \"config\": {\"markdown_headings\": true, \"markdown_code\": true, \"boilerplate_config\": {\"ratio_threshold\": 0.18, \"absolute_threshold\": 10, \"end_threshold\": 15, \"enable\": true}, \"remove_buttons\": true, \"remove_image_figures\": true, \"remove_link_clusters\": true, \"table_config\": {\"min_rows\": 2, \"min_cols\": 3, \"format\": \"plain\"}, \"remove_chinese\": true, \"remove_edit_buttons\": true, \"extract_latex\": true}, \"warc_path\": \"s3:\/\/commoncrawl\/crawl-data\/CC-MAIN-2019-13\/segments\/1552912204790.78\/warc\/CC-MAIN-20190326034712-20190326060712-00267.warc.gz\"}"}
| null | null |
ACCEPTED
#### According to
Index Fungorum
#### Published in
J. Proc. R. Soc. N. S. W. 46: 202 (1912)
#### Original name
Ustilago panici-gracilis E. Mackinnon
### Remarks
null
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 7,764
|
title: 'A gentle intro to Wix Installer'
date: '2016-08-31 19:13:00'
layout: single
author_profile: true
read_time: true
comments: true
share: true
related: true
---
There are a number of ways to create an installer for an application when it's ready to be shipped. Some of these ways are quite specific to the application and the language used to develop it. WiX Installer is one such approach, that has been around for quite some time (still in active development) can be used for pretty much anything on the Windows platform. What's great about it is that is has a boat load of features, including bootstrapping multiple installers together, running custom .NET scripts before or after certain events, and the ability to create repair/uninstall setups. WiX always outputs an MSI (.msi), so sys-admins prefer it over exe's.
Learning how to make a proper installer with WiX can be daunting. And the *here and there* documentation doesn't make it easy.
Here's a quick crash course to get going.
### getting started
WiX is a toolset. Download it and install it from their official website.
There are two main ways to work with WiX.
1 - Using the command line
This is the quicker approach and can be used to build an installer in a matter of minutes. Just install WiX Toolset, write your setup file in xml format (this will be a .wxs file) and use the WiX tools to compile and link your files.
I encountered one such installer at work, which was an old legacy installer that was passed down developers over the years. Every time the install logic needed to be changed, an unfortunate dev had to manually change the files referenced in the single (yes, single) .wxs file and make any changes to the logic. I followed suite the first time I had to work with WiX, because I didn't know better. The next time however, I decided to rewrite a new installer using the second approach outlined below.
2 - Using Visual Studio
Using VS is most certainly the recommended way for a number of reasons. It's easier to organize the setup files and imports, and will produce a faster workflow as it's possible to quickly compile and find any immediate errors in the files. If the application isn't a school project that's going to be deleted next week, this is the way to go. So first,
- Install Visual Studio
- Install WiX Toolset (must be done after installing VS if you want the build tools to be automatically installed)
- Install the WiX Toolset Build Tools for your version of VS
- In VS, create a new project like so: Create Project > Windows Installer XML > Setup Project
### structure of a project
What the hell are these file extensions? Let's break them down.
- .wxs files
Setup (configuration) files that WiX uses are in XML format with the file extension .wxs. These files will store what components (files and folders) are included in your installer, where they'll be installed, their folder structure as well as the interface shown by the installer. There is a naming convention where the main setup file is named Product.wxs. Place the 'components' in here, along with your application name, version, etc.
- .wixproj file
This file contains some meta data about your VS installer project. You can use this file to build your installer from the command line using MSBuild (used for CI, building without VS, needs .NET framework installed).
### creating a simple installer
1. First create a new installer project in VS. This results in a sample Product.wxs file.
2. Add a company name to the 'Manufacturer' attribute in the Product tag. The project can't build without this.
3. Add a new component to it so that the installer will actually install something. A single component usually corresponds to a single file.
To do this, first go into the project directory and create a new file there (eg: MyApp.txt). So this installer will installer this file to a users computer.
Now lets add this file as a component to your installer project so that it will be copied over to the users system during the installation process. To do this, simply add a new component inside the already existing ComponentGroup named 'ProductComponents'.
```
<Component Id='MainFile' Guid='*' Directory="INSTALLFOLDER">
<File Id='MyApp.txt' Name=' MyApp.txt' DiskId='1' Source='MyApp.txt' KeyPath='yes'/>
</Component>
```
Now you're ready to build the project. In a few seconds, the installer will be ready. Now go ahead and run the setup.
Note that there are no install configurations available to the user. It installs to the default location set in the code. Also note that the cab file (cab1.cab) is where data required by the installer is stored, and the setup cannot function without it. These files were needed back in the day when installers had to be separated so that parts could be shipped in different CD's or downloaded part by part. To fix this, find the MediaTemplate tag and add an EmbedCab attribute to it.
```
<MediaTemplate EmbedCab="yes" />
```
And the simple installer is done.
|
{
"redpajama_set_name": "RedPajamaGithub"
}
| 1,145
|
Дэниел Питер Грейвс (, род. 7 августа 1973 года) — профессиональный бейсболист, выступавший в Главной лиге бейсбола. Грейвс родился в Сайгоне в семье американского военного и вьетнамки и является единственным родившимся во Вьетнаме игроком в истории МЛБ и одним из нескольких вьетнамо-американских игроков. Большую часть профессиональной карьеры Грейвс провёл в «Цинциннати Редс», где с 1999 по 2004 год был лидером команды по количеству сейвов, за исключением сезона 2003 года, когда он выходил на позиции стартового питчера. На студенческом уровне выступал за команду университета Майами.
Профессиональная карьера
Грейвс был выбран на драфте МЛБ 2004 года в четвёртом раунде клубом «Кливленд Индианс». Уже через два дня после драфта, во время игр университетской Мировой серии Дэнни получил травму передней крестообразной связки. Вернувшись к выступлениям через год, он стал одним из лучших питчеров «Индианс» в низших лигах, а в 1996 году дебютировал в МЛБ. Проведя полтора года в Кливледне, в июле 1997 года Грейвс был обменян в «Цинциннати Редс».
За первые девять сезонов в МЛБ в составе «Индианс» и «Редс» Грейвс одержал 40 побед и потерпел 42 поражения. Он сделал 406 страйкаутов, 172 сейва, а его показатель ERA составил 3,89. Он является единственным игроком в истории МЛБ, у которого более чем в одном сезоне все хиты были хоум-ранами — в 2000 и 2001 годах (по 1 хоум-рану).
В 2003 году руководство «Редс» решило попробовать Грейвса на позиции стартового питчера. В итоге, за 26 стартов он показал результат 4-14.
Сезон 2005 года оказался ещё более неудачным для Дэнни. За первые 20 игр его показатель ERA составил 7,36. Такая невыразительная игра питчера пришлась не по нраву болельщикам «Редс», которые стали освистывать Дэнни каждый раз когда тот выходил на поле. Не выдержав напряжения, 23 мая, когда менеджер клуба Дэйв Майли вышел, чтобы заменить Грейвса, тот показал непристойный жест рукой болельщикам. Вскоре после этого поступка его отчислили из команды.
11 июня 2005 года он в качестве свободного агента подписал контракт с «Нью-Йорк Метс». В новой команде его игра по-прежнему оставляла желать лучшего и 26 августа его отправили в фарм-клуб, где он показал результат 0-2 с показателем ERA 18,0.
19 декабря 2005 года Грейвс подписал контракт с «Кливленд Индианс». Первоначально его планировали использовать в низших лигах, но отыграв хорошо весенние сборы Грейвс получил место в основном составе «Индианс». В играх же регулярного чемпионата Дэнни вновь стал испытывать проблемы и 18 мая 2006 года был переведён в фарм-клуб «Баффало Байзонс».
19 декабря 2006 года Грейвс подписал контракт на выступления в низших лигах с «Колорадо Рокиз», но ещё во время весенних сборов был отчислен. Чтобы поддерживать игровую практику Дэнни пришлось подписать контракт с «Лонг-Айленд Дакс» из Атлантической лиги профессионального бейсбола, в которой он стал лучшим по сейвам.
30 марта 2008 года Грейвс подписал контракт с «Миннесотой Твинс» и год выступал в фарм-клубе «Рочестер Уингс». По окончании сезона он стал свободным агентом и в январе 2009 года подписал контракт с «Хьюстон Астрос», но 25 марта был уволен.
Примечания
Ссылки
Vietnamese American Pitcher a Man for All Seasons
Danny Graves official website
Reds: Graves is still just a big kid
National Public Radio story about Graves «Bringing Baseball to Vietnam» — audio archive
Питчеры МЛБ
Игроки «Кливленд Индианс»
Игроки «Цинциннати Редс»
Игроки «Нью-Йорк Метс»
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 5,660
|
Jean Mannheim Plein Air Painting, 24" by 28"
Priced in US Dollars. This Plein Air painting was a gift from the artist to a neighbor in Pasadena in the early 1940's. It has stayed in that family uninterrupted to this day. The artist, Jean Mannheim, was born in Bad Kreuznach on the Nahe, Germany on Nov. 18, 1863. After being drafted into the German army, Mannheim deserted and fled to France where he studied art at Ecole Delecluse, Académie Colarossi, and with DeLancey and Bouguereau. Having learned book binding early in life, he used this trade to support himself while studying art in Paris. Upon immigrating to Illinois in 1884, he painted portraits in Chicago and taught in a Decatur art school. About 1903 he accepted a position at Frank Brangwyn's school in London and stayed for two years. Returning to the U.S., he taught at the Denver Art School until 1908. He then made his final move to Pasadena and built a home in the Arroyo Seco. Mannheim maintained a studio in the Blanchard Building in Los Angeles where he exhibited and taught, and in 1913 founded the Stickney Memorial School of Fine Arts in Pasadena. His figure studies and landscapes prior to 1915 were tighter and done with a restricted palette; whereas, his palette then lightened and he adopted the loose brushwork of Impressionism. He died in Pasadena on Sept. 6, 1945. Member: Laguna Beach AA; Long Beach AA. This measures 24" x 28" framed and is oil on masonite (a popular surface used by the artist) .
Free Delivery & Guaranteed Returns
Our free delivery zone includes all of the Lakeside from El Chante to Vista del Lago and all the metro Guadalajara area. Guaranteed returns within 7 days of your purchase.
PayPal for virtual/distanced orders.
Credit card, cash or check in store.
Cash or check upon delivery.
Payment made within 48 hours guarantees your purchase. Items will be sold to the next interested person if payment is not made.
FREE DELIVERY from Chapala to West Ajijic (El Bajio). Extends to all of Lakeside and metro Guadalajara if total purchase is over $4000 pesos.
SMALL DELIVERY CHARGE for all other areas of Lakeside if total purchase is under $2000 pesos.
PICK UP & DELIVERY hours from 4 to 6 p.m. Tuesday through Friday.
7 DAY GUARANTEED returns for all purchases made through the website.
Thank you, we will be in touch soon!
Address: Independencia 7, 45920 Ajijic, Jalisco, Mexico
Phone: +52 37 6766 3636, +52 33 1497 3620
Hours: Tuesday through Saturday 10 a.m. - 2 p.m.
Closed Sunday and Mondays
Copyright © Barbara's Bazaar/Ajijic4me.com Web Design by Rising By Design
|
{
"redpajama_set_name": "RedPajamaCommonCrawl"
}
| 7,150
|
This segment reveals the results of a study on buying the /VX or /VX covered calls to determine if there is a mechanical strategy that works. The results of the study should be of interest to all traders. A bonus examination of market moves around FOMC meetings precedes the main topic.
A graph of the VIX was displayed. We see how the VIX occasionally spikes from low levels. How to benefit from this was discussed. One way would be to buy the /VX and using options one could make it a covered call. An example was shown.
A study was conducted using VIX futures and options from June 2008 to present. On the day of each VIX settlement, we bought a /VX future and sold 10 of the first out-of-money (OTM) VIX calls in the contract with 27-34 days until settlement. The usual 45 day target was not used because the liquidity is in the front month contract.
A table was displayed of all 88 trades, being long the /VX (VIX Future) as well as a /VX covered call. The table included the total P/L, average P/L, percentage of profitability, largest profit and largest loss.
A second table was displayed of all 32 trades when the /VX was trading below 17.50 The table included being long the /VX as well as a /VX covered call. The table showed the total P/L, average P/L, percentage of profitability, largest profit and largest loss.
A third table was displayed of all 16 trades when the /VX was trading below 15.00 The table included being long the /VX as well as a /VX covered call. The table showed the total P/L, average P/L, percentage of profitability, largest profit and largest loss.
Watch this segment of "Market Measures" with Tom Sosnoff and Tony Battista for the takeaways, the study results and other important information about understanding and trading /VX (VIX Future) as well as /VX covered calls in different volatility environments.
|
{
"redpajama_set_name": "RedPajamaC4"
}
| 4,707
|
Jesús Sánchez was a Mexican fencer. He competed in the team foil event at the 1932 Summer Olympics.
References
External links
Year of birth missing
Year of death missing
Mexican male foil fencers
Olympic fencers of Mexico
Fencers at the 1932 Summer Olympics
|
{
"redpajama_set_name": "RedPajamaWikipedia"
}
| 8,480
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.